Core-selecting combinatorial auctions are popular auction designs that constrain prices to eliminate the incentive for any group of bidders---with the seller---to renegotiate for a better deal. They help overcome the low-revenue issues of classical combinatorial auctions. We introduce a new class of core-selecting combinatorial auctions that leverage bidder information available to the auction designer. We model such information through constraints on the joint type space of the bidders---these are constraints on bidders' private valuations that are known to hold by the auction designer before bids are elicited. First, we show that type space information can overcome the well-known impossibility of incentive-compatible core-selecting combinatorial auctions. We present a revised and generalized version of that impossibility result that depends on how much information is conveyed by the type spaces. We then devise a new family of core-selecting combinatorial auctions and show that they minimize the sum of bidders' incentives to deviate from truthful bidding. We develop new constraint generation techniques---and build upon existing quadratic programming techniques---to compute core prices, and conduct experiments to evaluate the incentive, revenue, fairness, and computational merits of our new auctions. Our new core-selecting auctions directly improve upon existing designs that have been used in many high-stakes auctions around the world. We envision that they will be a useful addition to any auction designer's toolkit.

Counterfactual regret minimization (CFR) algorithms are a foundational class of methods for solving imperfect-information games, with the time average of their iterates converging to a Nash equilibrium in two-player zero-sum games. Prior state-of-the-art variants, Discounted CFR (DCFR) and Predictive CFR+ (PCFR+), achieved the fastest known practical performance by improving convergence rates over vanilla CFR through discounting early iterations with a fixed discounting scheme. More recently, Dynamic DCFR (DDCFR) introduced agent-learned dynamic discounting schemes to further accelerate convergence, at the cost of increased complexity. To address this, we propose Hyperparameter Schedules (HSs), a remarkably simple, training-free framework that dynamically adjusts CFR discounting over time. HSs aggressively downweight early updates and gradually transition to trusting late-stage strategies, leading to substantially faster convergence with only a few lines of code modifications. We show that HSs derived from just three small extensive-form games generalize effectively to 17 diverse games (including large-scale realistic poker) in both extensive-form and normal-form settings, without any game-specific tuning. Our method establishes a new state of the art for solving two-player zero-sum games.

The design of multi-item, multi-bidder auctions involves a delicate balancing act of economic objectives, bidder incentives, and real-world complexities. Efficient auctions, that is, auctions that allocate items to maximize total bidder value, are practically desirable since they promote the most economically beneficial use of resources. Arguably the biggest drawback of efficient auctions, however, is their potential to generate very low revenue. In this work, we show how the auction designer can artificially inject competition into the auction to boost revenue while striving to maintain efficiency. First, we invent a new auction family that enables the auction designer to specify competition in a precise, expressive, and interpretable way. We then introduce a new model of bidder behavior and individual rationality to understand how bidders act when prices are too competitive. Next, under our bidder behavior model, we use our new competitive auction class to derive the globally revenue-optimal efficient auction under two different knowledge models for the auction designer: knowledge of full bidder value distributions and knowledge of bidder value quantiles. Finally, we study a third knowledge model for the auction designer: knowledge of historical bidder valuation data. In this setting we present sample and computationally efficient learning algorithms that find high-revenue probably-efficient competitive auctions from bidder data. Our learning algorithms are instance adaptive and can be run in parallel across bidders, unlike most prior approaches to data-driven auction design.

Imperfect-recall games—in which players may forget previously acquired information—have found many practical applications, ranging from game abstractions to team games and testing AI agents. In this paper, we quantify the utility gain by endowing a player with perfect recall, which we call the value of recall (VoR). While VoR can be unbounded in general, we parameterize it in terms of various game properties, namely the structure of chance nodes and the degree of absentmindedness (the number of successive times a player enters the same information set). Further, we identify several pathologies that arise with VoR, and show how to circumvent them. We also study the complexity of computing VoR, and how to optimally apportion partial recall. Finally, we connect VoR to other previously studied concepts in game theory, including the price of anarchy. We use that connection in conjunction with the celebrated smoothness framework to characterize VoR in a broad class of games.

International audience

Strategic interactions can be represented more concisely, and analyzed and solved more efficiently, if we are aware of the symmetries within the multiagent system. Symmetries also have conceptual implications, for example for equilibrium selection. We study the computational complexity of identifying and using symmetries. Using the classical framework of normal-form games, we consider game symmetries that can be across some or all players and/or actions. We find a strong connection between game symmetries and graph automorphisms, yielding graph automorphism and graph isomorphism completeness results for characterizing the symmetries present in a game. On the other hand, we also show that the problem becomes polynomial-time solvable when we restrict the consideration of actions in one of two ways. Next, we investigate when exactly game symmetries can be successfully leveraged for Nash equilibrium computation. We show that finding a Nash equilibrium that respects a given set of symmetries is PPAD- and CLS-complete in general-sum and team games respectively---that is, exactly as hard as Brouwer fixed point and gradient descent problems. Finally, we present polynomial-time methods for the special cases where we are aware of a vast number of symmetries, or where the game is two-player zero-sum and we do not even know the symmetries.

*Variational inequalities (VIs)* encompass many fundamental problems in diverse areas ranging from engineering to economics and machine learning. However, their considerable expressivity comes at the cost of computational intractability. In this paper, we introduce and analyze a natural relaxation—which we refer to as *expected variational inequalities (EVIs)*—where the goal is to find a distribution that satisfies the VI constraint in expectation. By adapting recent techniques from game theory, we show that, unlike VIs, EVIs can be solved in polynomial time under general (nonmonotone) operators. EVIs capture the seminal notion of correlated equilibria, but enjoy a greater reach beyond games. We also employ our framework to capture and generalize several existing disparate results, including from settings such as smooth games, and games with coupled constraints or nonconcave utilities.

We study the problem of computing an approximate Nash equilibrium of a game whose strategy space is continuous without access to gradients of the utility function. Lack of access to gradients is common in reinforcement learning settings, where the environment is treated as a black box, as well as equilibrium finding in mechanisms such as auctions, where the mechanism's payoffs are discontinuous in the players' actions. To tackle this problem, we turn to zeroth-order optimization techniques that combine pseudo-gradients with equilibrium-finding dynamics. Specifically, we introduce a new technique that requires a number of utility function evaluations per iteration that is constant rather than linear in the number of players. It achieves this by performing a single joint perturbation on all players' strategies, rather than perturbing each one individually. This is very important for many-player games, especially when the utility function is expensive to compute in terms of wall time, memory, money, or other resources. We evaluate our approach on various games, including auctions, which have important real-world applications. Our approach yields a dramatic improvement in performance in terms of the wall time required to reach an approximate Nash equilibrium.

Sequence-independent lifting is a procedure for strengthening valid inequalities of an integer program. We generalize the sequence-independent lifting method of Gu, Nemhauser, and Savelsbergh (GNS lifting) for cover inequalities and correct an error in their proposed generalization. We obtain a new sequence-independent lifting technique---piecewise-constant (PC) lifting---with a number of important properties. We derive a broad set of sufficient conditions under which PC lifting yields facets---the first characterization of facet-defining sequence-independent liftings that are efficiently computable from the underlying cover. Finally, we demonstrate via experiments that PC lifting can be a useful alternative to GNS lifting. We test PC lifting atop a number of novel cover inequality generation routines, which prove to be effective in experiments with CPLEX. PC lifting delivers strong numerical properties making it practically relevant for integer programming solvers.

We consider the problem of computing stationary points in min-max optimization, with a focus on the special case of Nash equilibria in (two-)team zero-sum games. We first show that computing $\epsilon$-Nash equilibria in $3$-player $\text{\emph{adversarial}}$ team games---wherein a team of $2$ players competes against a $\text{\emph{single}}$ adversary---is $\textsf{CLS}$-complete, resolving the complexity of Nash equilibria in such settings. Our proof proceeds by reducing from $\text{\emph{symmetric}}$ $\epsilon$-Nash equilibria in $\text{\emph{symmetric}}$, identical-payoff, two-player games, by suitably leveraging the adversarial player so as to enforce symmetry---without disturbing the structure of the game. In particular, the class of instances we construct comprises solely polymatrix games, thereby also settling a question left open by Hollender, Maystre, and Nagarajan (2024). Moreover, we establish that computing $\text{\emph{symmetric}}$ (first-order) equilibria in $\text{\emph{symmetric}}$ min-max optimization is $\textsf{PPAD}$-complete, even for quadratic functions. Building on this reduction, we show that computing symmetric $\epsilon$-Nash equilibria in symmetric, $6$-player ($3$ vs. $3$) team zero-sum games is also $\textsf{PPAD}$-complete, even for $\epsilon = \text{poly}(1/n)$. As a corollary, this precludes the existence of symmetric dynamics---which includes many of the algorithms considered in the literature---converging to stationary points. Finally, we prove that computing a $\text{\emph{non-symmetric}}$ $\text{poly}(1/n)$-equilibrium in symmetric min-max optimization is $\textsf{FNP}$-hard.

Φ-equilibria—and the associated notion of Φ-regret—are a powerful and flexible framework at the heart of online learning and game theory, whereby enriching the set of deviations Φ begets stronger notions of rationality. Recently, Daskalakis, Farina, Fishelson, Pipis, and Schneider (STOC '24) settled the existence of efficient algorithms when Φ contains only linear maps under a general, d-dimensional convex constraint set X. In this paper, we significantly extend their work by resolving the case where Φ is k-dimensional; degree-ℓ polynomials constitute a canonical such example with k = dO(ℓ). In particular, we obtain two main positive results: i) a poly(n, d, k, log(1/ϵ))-time algorithm for computing ϵ-approximate Φ-equilibria in n-player multilinear games, and ii) an efficient online algorithm that incurs average Φ-regret at most ϵ using poly(d, k)/ϵ2 rounds.

Policy gradient methods enjoy strong practical performance in numerous tasks in reinforcement learning. Their theoretical understanding in multiagent settings, however, remains limited, especially beyond two-player competitive and potential Markov games. In this paper, we develop a new framework to characterize optimistic policy gradient methods in multi-player Markov games with a single controller. Specifically, under the further assumption that the game exhibits an equilibrium collapse, in that the marginals of coarse correlated equilibria (CCE) induce Nash equilibria (NE), we show convergence to stationary epsilon-NE in O(1/epsilon^2) iterations, where O suppresses polynomial factors in the natural parameters of the game. Such an equilibrium collapse is well-known to manifest itself in two-player zero-sum Markov games, but also occurs even in a class of multi-player Markov games with separable interactions, as established by recent work. As a result, we bypass known complexity barriers for computing stationary NE when either of our assumptions fails. Our approach relies on a natural generalization of the classical Minty property that we introduce, which we anticipate to have further applications beyond Markov games.

Dynamic mechanism design is a challenging extension to ordinary mechanism design in which the mechanism designer must make a sequence of decisions over time in the face of possibly untruthful reports of participating agents. Optimizing dynamic mechanisms for welfare is relatively well understood. However, there has been less work on optimizing for other goals (e.g., revenue), and without restrictive assumptions on valuations, it is remarkably challenging to characterize good mechanisms. Instead, we turn to automated mechanism design to find mechanisms with good performance in specific problem instances. We extend the class of affine maximizer mechanisms to MDPs where agents may untruthfully report their rewards. This extension results in a challenging bilevel optimization problem in which the upper problem involves choosing optimal mechanism parameters, and the lower problem involves solving the resulting MDP. Our approach can find truthful dynamic mechanisms that achieve strong performance on goals other than welfare, and can be applied to essentially any problem setting---without restrictions on valuations---for which RL can learn optimal policies.

We investigate two notions of correlated equilibrium for extensive-form games: the extensive-form correlated equilibrium (EFCE) and the behavioral correlated equilibrium (BCE). We show that the two are outcome-equivalent, in the sense that every outcome distribution achievable under one notion is achievable under the other. Our result implies, to our knowledge, the first polynomial-time algorithm for computing a BCE.

Deploying reinforcement learning (RL) systems requires robustness to uncertainty and model misspecification, yet prior robust RL methods typically only study noise introduced independently across time. However, practical sources of uncertainty are usually coupled across time. We formally introduce temporally-coupled perturbations, presenting a novel challenge for existing robust RL methods. To tackle this challenge, we propose GRAD, a novel game-theoretic approach that treats the temporally-coupled robust RL problem as a partially-observable two-player zero-sum game. By finding an approximate equilibrium within this game, GRAD optimizes for general robustness against temporally-coupled perturbations. Experiments on continuous control tasks demonstrate that, compared with prior methods, our approach achieves a higher degree of robustness to various types of attacks on different attack domains, both in settings with temporally-coupled perturbations and decoupled perturbations.

In competitive two-agent environments, deep reinforcement learning (RL) methods like Policy Space Response Oracles (PSRO) often increase exploitability between iterations, which is problematic when training in large games. To address this issue, we introduce anytime double oracle (ADO), an algorithm that ensures exploitability does not increase between iterations, and its approximate extensive-form version, anytime PSRO (APSRO). ADO converges to a Nash equilibrium while iteratively reducing exploitability. However, convergence in these algorithms may require adding all of a game's deterministic policies. To improve this, we propose Self-Play PSRO (SP-PSRO), which incorporates an approximately optimal stochastic policy into the population in each iteration. APSRO and SP-PSRO demonstrate lower exploitability and near-monotonic exploitability reduction in games like Leduc poker and Liar's Dice. Empirically, SP-PSRO often converges much faster than APSRO and PSRO, requiring only a few iterations in many games.

Large language models are typically aligned with human preferences by optimizing reward models (RMs) fitted to human feedback. However, human preferences are multi-faceted, and it is increasingly common to derive reward from a composition of simpler reward models which each capture a different aspect of language quality. This itself presents a challenge, as it is difficult to appropriately weight these component RMs when combining them. Compounding this difficulty, because any RM is only a proxy for human evaluation, this process is vulnerable to *overoptimization*, wherein past a certain point, accumulating higher reward is associated with worse human ratings. In this paper, we perform the first study on overoptimization in composite RMs, showing that correlation between component RMs has a significant effect on the locations of these points. We then introduce an approach to solve this issue using constrained reinforcement learning as a means of preventing the agent from exceeding each RM's threshold of usefulness. Our method addresses the problem of weighting component RMs by learning dynamic weights, naturally given by the Lagrange multipliers. As a result, each RM stays within the range at which it is an effective proxy, improving evaluation performance. Finally, we introduce an adaptive method using gradient-free optimization to identify and optimize towards these points during a single run.

A recent paper by Farina and Pipis (2023) established the existence of uncoupled no-linear-swap regret dynamics with polynomial-time iterations in extensive-form games. The equilibrium points reached by these dynamics, known as linear correlated equilibria, are currently the tightest known relaxation of correlated equilibrium that can be learned in polynomial time in any finite extensive-form game. However, their properties remain vastly unexplored, and their computation is onerous. In this paper, we provide several contributions shedding light on the fundamental nature of linear-swap regret. First, we show a connection between linear deviations and a generalization of communication deviations in which the player can make queries to a ``mediator'' who replies with action recommendations, and, critically, the player is not constrained to match the timing of the game as would be the case for communication deviations. We coin this latter set the untimed communication (UTC) deviations. We show that the UTC deviations coincide precisely with the linear deviations, and therefore that any player minimizing UTC regret also minimizes linear-swap regret. We then leverage this connection to develop state-of-the-art no-regret algorithms for computing linear correlated equilibria, both in theory and in practice. In theory, our algorithms achieve polynomially better per-iteration runtimes; in practice, our algorithms represent the state of the art by several orders of magnitude.

Gradient-based algorithms have shown great promise in solving large (two-player) zero-sum games. However, their success has been mostly confined to the low-precision regime since the number of iterations grows polynomially in $1/\epsilon$, where $\epsilon > 0$ is the duality gap. While it has been well-documented that linear convergence---an iteration complexity scaling as $\text{log}(1/\epsilon)$---can be attained even with gradient-based algorithms, that comes at the cost of introducing a dependency on certain condition number-like quantities which can be exponentially large in the description of the game. To address this shortcoming, we examine the iteration complexity of several gradient-based algorithms in the celebrated framework of smoothed analysis, and we show that they have polynomial smoothed complexity, in that their number of iterations grows as a polynomial in the dimensions of the game, $\text{log}(1/\epsilon)$, and $1/\sigma$, where $\sigma$ measures the magnitude of the smoothing perturbation. Our result applies to optimistic gradient and extra-gradient descent/ascent, as well as a certain iterative variant of Nesterov's smoothing technique. From a technical standpoint, the proof proceeds by characterizing and performing a smoothed analysis of a certain error bound, the key ingredient driving linear convergence in zero-sum games. En route, our characterization also makes a natural connection between the convergence rate of such algorithms and perturbation-stability properties of the equilibrium, which is of interest beyond the model of smoothed complexity.

Recent breakthrough results by Dagan, Daskalakis, Fishelson and Golowich [2023] and Peng and Rubinstein [2023] established an efficient algorithm attaining at most $\epsilon$ swap regret over extensive-form strategy spaces of dimension $N$ in $N^{\tilde O(1/\epsilon)}$ rounds. On the other extreme, Farina and Pipis [2023] developed an efficient algorithm for minimizing the weaker notion of linear-swap regret in $\mathsf{poly}(N)/\epsilon^2$ rounds. In this paper, we develop efficient parameterized algorithms for regimes between these two extremes. We introduce the set of $k$-mediator deviations, which generalize the untimed communication deviations recently introduced by Zhang, Farina and Sandholm [2024] to the case of having multiple mediators, and we develop algorithms for minimizing the regret with respect to this set of deviations in $N^{O(k)}/\epsilon^2$ rounds. Moreover, by relating $k$-mediator deviations to low-degree polynomials, we show that regret minimization against degree-$k$ polynomial swap deviations is achievable in $N^{O(kd)^3}/\epsilon^2$ rounds, where $d$ is the depth of the game, assuming a constant branching factor. For a fixed degree $k$, this is polynomial for Bayesian games and quasipolynomial more broadly when $d = \mathsf{polylog} N$---the usual balancedness assumption on the game tree. The first key ingredient in our approach is a relaxation of the usual notion of a fixed point required in the framework of Gordon, Greenwald and Marks [2008]. Namely, for a given deviation $\phi$, we show that it suffices to compute what we refer to as a fixed point in expectation; that is, a distribution $\pi$ such that $\mathbb{E}_{x \sim \pi} [\phi(x) - x] \approx 0$. Unlike the problem of computing an actual (approximate) fixed point $x \approx \phi(x)$, which we show is \PPAD-hard, there is a simple and efficient algorithm for finding a solution that satisfies our relaxed notion. As a byproduct, we provide, to our knowledge, the fastest algorithm for computing $\epsilon$-correlated equilibria in normal-form games in the medium-precision regime, obviating the need to solve a linear system in every round. Our second main contribution is a characterization of the set of low-degree deviations, made possible through a connection to low-depth decisions trees from Boolean analysis.

In this paper, we study the class of games known as hidden-role games in which players are assigned privately to teams and are faced with the challenge of recognizing and cooperating with teammates. This model includes both popular recreational games such as the Mafia/Werewolf family and The Resistance (Avalon) and many real-world settings, such as distributed systems where nodes need to work together to accomplish a goal in the face of possible corruptions. There has been little to no formal mathematical grounding of such settings in the literature, and it was previously not even clear what the right solution concepts (notions of equilibria) should be. A suitable notion of equilibrium should take into account the communication channels available to the players (e.g., can they communicate? Can they communicate in private?). Defining such suitable notions turns out to be a nontrivial task with several surprising consequences. In this paper, we provide the first rigorous definition of equilibrium for hidden-role games, which overcomes serious limitations of other solution concepts not designed for hidden-role games. We then show that in certain cases, including the above recreational games, optimal equilibria can be computed efficiently. In most other cases, we show that computing an optimal equilibrium is at least NP-hard or coNP-hard. Lastly, we experimentally validate our approach by computing exact equilibria for complete 5- and 6-player Avalon instances whose size in terms of number of information sets is larger than 1056.

A mediator observes no-regret learners playing an extensive-form game repeatedly across T rounds. The mediator attempts to steer players toward some desirable predetermined equilibrium by giving (nonnegative) payments to players. We call this the steering problem. The steering problem captures problems several problems of interest, among them equilibrium selection and information design (persuasion). If the mediator's budget is unbounded, steering is trivial because the mediator can simply pay the players to play desirable actions. We study two bounds on the mediator's payments: a total budget and a per-round budget. If the mediator's total budget does not grow with T, we show that steering is impossible. However, we show that it is enough for the total budget to grow sublinearly with T, that is, for the average payment to vanish. When players' full strategies are observed at each round, we show that constant per-round budgets permit steering. In the more challenging setting where only trajectories through the game tree are observable, we show that steering is impossible with constant per-round budgets in general extensive-form games, but possible in normal-form games or if the per-round budget may itself depend on T. We also show how our results can be generalized to the case when the equilibrium is being computed online while steering is happening. We supplement our theoretical positive results with experiments highlighting the efficacy of steering in large games.

Algorithms often have tunable parameters that impact performance metrics such as runtime and solution quality. For many algorithms used in practice, no parameter settings admit meaningful worst-case bounds, so the parameters are made available for the user to tune. Alternatively, parameters may be tuned implicitly within the proof of a worst-case approximation ratio or runtime bound. Worst-case instances, however, may be rare or nonexistent in practice. A growing body of research has demonstrated that a data-driven approach to parameter tuning can lead to significant improvements in performance. This approach uses a training set of problem instances sampled from an unknown, application-specific distribution and returns a parameter setting with strong average performance on the training set. We provide techniques for deriving generalization guarantees that bound the difference between the algorithm’s average performance over the training set and its expected performance on the unknown distribution. Our results apply no matter how the parameters are tuned, be it via an automated or manual approach. The challenge is that for many types of algorithms, performance is a volatile function of the parameters: slightly perturbing the parameters can cause a large change in behavior. Prior research [e.g., 12 , 16 , 20 , 62 ] has proved generalization bounds by employing case-by-case analyses of greedy algorithms, clustering algorithms, integer programming algorithms, and selling mechanisms. We streamline these analyses with a general theorem that applies whenever an algorithm’s performance is a piecewise-constant, piecewise-linear, or—more generally— piecewise-structured function of its parameters. Our results, which are tight up to logarithmic factors in the worst case, also imply novel bounds for configuring dynamic programming algorithms from computational biology.

Tree search algorithms, such as branch-and-bound, are the most widely used tools for solving combinatorial and non-convex problems. For example, they are the foremost method for solving (mixed) integer programs and constraint satisfaction problems. Tree search algorithms come with a variety of tunable parameters that are notoriously challenging to tune by hand. A growing body of research has demonstrated the power of using a data-driven approach to automatically optimize the parameters of tree search algorithms. These techniques use a training set of integer programs sampled from an application-specific instance distribution to find a parameter setting that has strong average performance over the training set. However, with too few samples, a parameter setting may have strong average performance on the training set but poor expected performance on future integer programs from the same application. Our main contribution is to provide the first sample complexity guarantees for tree search parameter tuning. These guarantees bound the number of samples sufficient to ensure that the average performance of tree search over the samples nearly matches its future expected performance on the unknown instance distribution. In particular, the parameters we analyze weight scoring rules used for variable selection. Proving these guarantees is challenging because tree size is a volatile function of these parameters: we prove that, for any discretization (uniform or not) of the parameter space, there exists a distribution over integer programs such that every parameter setting in the discretization results in a tree with exponential expected size, yet there exist parameter settings between the discretized points that result in trees of constant size. In addition, we provide data-dependent guarantees that depend on the volatility of these tree-size functions: our guarantees improve if the tree-size functions can be well approximated by simpler functions. Finally, via experiments, we illustrate that learning an optimal weighting of scoring rules reduces tree size.

In this paper, we establish efficient and uncoupled learning dynamics so that, when employed by all players in multiplayer perfect-recall imperfect-information extensive-form games, the trigger regret of each player grows as $O(\log T)$ after $T$ repetitions of play. This improves exponentially over the prior best known trigger-regret bound of $O(T^{1/4})$, and settles a recent open question by Bai et al. (2022). As an immediate consequence, we guarantee convergence to the set of extensive-form correlated equilibria and coarse correlated equilibria at a near-optimal rate of $\frac{\log T}{T}$. Building on prior work, at the heart of our construction lies a more general result regarding fixed points deriving from rational functions with polynomial degree, a property that we establish for the fixed points of (coarse) trigger deviation functions. Moreover, our construction leverages a refined regret circuit for the convex hull, which—unlike prior guarantees—preserves the RVU property introduced by Syrgkanis et al. (NIPS, 2015); this observation has an independent interest in establishing near-optimal regret under learning dynamics based on a CFR-type decomposition of the regret.

A classic result in the theory of extensive-form games asserts that the set of strategies available to any perfect-recall player is strategically equivalent to a low-dimensional convex polytope, called the *sequence-form polytope*. Online convex optimization tools operating on this polytope are the current state-of-the-art for computing several notions of equilibria in games, and have been crucial in landmark applications of computational game theory. However, when optimizing over the *joint* strategy space of a *team* of players, one cannot use the sequence form to obtain a strategically-equivalent convex description of the strategy set of the team. In this paper, we provide new complexity results on the computation of optimal strategies for teams, and propose a new representation, coined *team belief DAG (TB-DAG)*, that describes team strategies as a convex set. The TB-DAG enjoys state-of-the-art parameterized complexity bounds, while at the same time enjoying the advantages of efficient regret minimization techniques. We show that TB-DAG can be exponentially smaller and can be computed exponentially faster than all other known representations, and that the converse is never true. Experimentally, we show that the TB-DAG, when paired with learning techniques, yields state of the art on a wide variety of benchmark team games.

A recent approach to automated mechanism design, differentiable economics, represents auctions by rich function approximators and optimizes their performance by gradient descent. The ideal auction architecture for differentiable economics would be perfectly strategyproof, support multiple bidders and items, and be rich enough to represent the optimal (i.e. revenue-maximizing) mechanism. So far, such an architecture does not exist. There are single-bidder approaches (MenuNet, RochetNet) which are always strategyproof and can represent optimal mechanisms. RegretNet is multi-bidder and can approximate any mechanism, but is only approximately strategyproof. We present an architecture that supports multiple bidders and is perfectly strategyproof, but cannot necessarily represent the optimal mechanism. This architecture is the classic affine maximizer auction (AMA), modified to offer lotteries. By using the gradient-based optimization tools of differentiable economics, we can now train lottery AMAs, competing with or outperforming prior approaches in revenue.

We study the problem of computing an approximate Nash equilibrium of continuous-action game without access to gradients. Such game access is common in reinforcement learning settings, where the environment is typically treated as a black box. To tackle this problem, we apply zeroth-order optimization techniques that combine smoothed gradient estimators with equilibrium-finding dynamics. We model players' strategies using artificial neural networks. In particular, we use randomized policy networks to model mixed strategies. These take noise in addition to an observation as input and can flexibly represent arbitrary observation-dependent, continuous-action distributions. Being able to model such mixed strategies is crucial for tackling continuous-action games that lack pure-strategy equilibria. We evaluate the performance of our method using an approximation of the Nash convergence metric from game theory, which measures how much players can benefit from unilaterally changing their strategy. We apply our method to continuous Colonel Blotto games, single-item and multi-item auctions, and a visibility game. The experiments show that our method can quickly find a high-quality approximate equilibrium. Furthermore, they show that the dimensionality of the input noise is crucial for performance. To our knowledge, this paper is the first to solve general continuous-action games with unrestricted mixed strategies and without any gradient information.

Coalition formation is a central approach for multiagent coordination. A crucial part of coalition formation that is extensively studied in AI is coalition structure generation: partitioning agents into coalitions to maximize overall value. In this paper, we propose a novel method for coalition structure generation by introducing a compact and efficient representation of coalition structures. Our representation partitions the solution space into smaller, more manageable subspaces that gather structures containing coalitions of specific sizes. Our proposed method combines two new algorithms, one which leverages our compact representation and a branch-and-bound technique to generate optimal coalition structures, and another that utilizes a preprocessing phase to identify the most promising sets of coalitions to evaluate. Additionally, we show how parts of the solution space can be gathered into groups to avoid their redundant evaluation and we investigate the computational gain that is achieved by avoiding that redundant processing. Through this approach, our algorithm is able to prune the solution space more efficiently. Our results show that the proposed algorithm is superior to prior state-of-the-art methods in generating optimal coalition structures under several value distributions.

Most of the literature on learning in games has focused on the restrictive setting where the underlying repeated game does not change over time. Much less is known about the convergence of no-regret learning algorithms in dynamic multiagent settings. In this paper, we characterize the convergence of optimistic gradient descent (OGD) in time-varying games. Our framework yields sharp convergence bounds for the equilibrium gap of OGD in zero-sum games parameterized on natural variation measures of the sequence of games, subsuming known results for static games. Furthermore, we establish improved second-order variation bounds under strong convexity-concavity, as long as each game is repeated multiple times. Our results also apply to time-varying general-sum multi-player games via a bilinear formulation of correlated equilibria, which has novel implications for meta-learning and for obtaining refined variation-dependent regret bounds, addressing questions left open in prior papers. Finally, we leverage our framework to also provide new insights on dynamic regret guarantees in static games.

Computational tractability and social welfare (aka. efficiency) of equilibria are two fundamental but in general orthogonal considerations in algorithmic game theory. Nevertheless, we show that when (approximate) full efficiency can be guaranteed via a smoothness argument a la Roughgarden, Nash equilibria are approachable under a family of no-regret learning algorithms, thereby enabling fast and decentralized computation. We leverage this connection to obtain new convergence results in large games---wherein the number of players $n \gg 1$---under the well-documented property of full efficiency via smoothness in the limit. Surprisingly, our framework unifies equilibrium computation in disparate classes of problems including games with vanishing strategic sensitivity and two-player zero-sum games, illuminating en route an immediate but overlooked equivalence between smoothness and a well-studied condition in the optimization literature known as the Minty property. Finally, we establish that a family of no-regret dynamics attains a welfare bound that improves over the smoothness framework while at the same time guaranteeing convergence to the set of coarse correlated equilibria. We show this by employing the clairvoyant mirror descent algortihm recently introduced by Piliouras et al.

Recent algorithms have achieved superhuman performance at a number of two-player zero-sum games such as poker and go. However, many real-world situations are multi-player games. Zero-sum two-team games, such as bridge and football, involve two teams where each member of the team shares the same reward with every other member of that team, and each team has the negative of the reward of the other team. A popular solution concept in this setting, called TMECor, assumes that teams can jointly correlate their strategies before play, but are not able to communicate during play. This setting is harder than two-player zero-sum games because each player on a team has different information and must use their public actions to signal to other members of the team. Prior works either have game-theoretic guarantees but only work in very small games, or are able to scale to large games but do not have game-theoretic guarantees. In this paper we introduce two algorithms: Team-PSRO, an extension of PSRO from two-player games to team games, and Team-PSRO Mix-and-Match which improves upon Team PSRO by better using population policies. In Team-PSRO, in every iteration both teams learn a joint best response to the opponent's meta-strategy via reinforcement learning. As the reinforcement learning joint best response approaches the optimal best response, Team-PSRO is guaranteed to converge to a TMECor. In experiments on Kuhn poker and Liar's Dice, we show that a tabular version of Team-PSRO converges to TMECor, and a version of Team PSRO using deep cooperative reinforcement learning beats self-play reinforcement learning in the large game of Google Research Football.

We develop a versatile new methodology for multidimensional mechanism design that incorporates side information about agent types to generate high social welfare and high revenue simultaneously. Prominent sources of side information in practice include predictions from a machine-learning model trained on historical agent data, advice from domain experts, and even the mechanism designer's own gut instinct. In this paper we adopt a prior-free perspective that makes no assumptions on the correctness, accuracy, or source of the side information. First, we design a meta-mechanism that integrates input side information with an improvement of the classical VCG mechanism. The welfare, revenue, and incentive properties of our meta-mechanism are characterized by novel constructions we introduce based on the notion of a weakest competitor, which is an agent that has the smallest impact on welfare. We show that our meta-mechanism, when carefully instantiated, simultaneously achieves strong welfare and revenue guarantees parameterized by errors in the side information. When the side information is highly informative and accurate, our mechanism achieves welfare and revenue competitive with the total social surplus, and its performance decays continuously and gradually as the quality of the side information decreases. Finally, we apply our meta-mechanism to a setting where each agent's type is determined by a constant number of parameters. Specifically, agent types lie on constant-dimensional subspaces (of the potentially high-dimensional ambient type space) that are known to the mechanism designer. We use our meta-mechanism to obtain the first known welfare and revenue guarantees in this setting.

We introduce a new approach for computing optimal equilibria via learning in games. It applies to extensive-form settings with any number of players, including mechanism design, information design, and solution concepts such as correlated, communication, and certification equilibria. We observe that optimal equilibria are minimax equilibrium strategies of a player in an extensive-form zero-sum game. This reformulation allows to apply techniques for learning in zero-sum games, yielding the first learning dynamics that converge to optimal equilibria, not only in empirical averages, but also in iterates. We demonstrate the practical scalability and flexibility of our approach by attaining state-of-the-art performance in benchmark tabular games, and by computing an optimal mechanism for a sequential auction design problem using deep reinforcement learning.

Despite the many recent practical and theoretical breakthroughs in computational game theory, equilibrium finding in extensive-form team games remains a significant challenge. While NP-hard in the worst case, there are provably efficient algorithms for certain families of team game. In particular, if the game has common external information, also known as A-loss recall---informally, actions played by non-team members (i.e., the opposing team or nature) are either unknown to the entire team, or common knowledge within the team---then polynomial-time algorithms exist. In this paper, we devise a completely new algorithm for solving team games. It uses a tree decomposition of the constraint system representing each team's strategy to reduce the number and degree of constraints required for correctness (tightness of the mathematical program). Our approach has the bags of the tree decomposition correspond to team-public states---that is, minimal sets of nodes (that is, states of the team) such that, upon reaching the set, it is common knowledge among the players on the team that the set has been reached. Our algorithm reduces the problem of solving team games to a linear program with at most O(NW^(w+1)) nonzero entries in the constraint matrix, where N is the size of the game tree, w is a parameter that depends on the amount of uncommon external information, and W is the treewidth of the tree decomposition. In public-action games, our program size is bounded by the tighter 2^(O(nt))N for teams of n players with t types each. Our algorithm is based on a new way to write a custom, concise tree decomposition, and its fast run time does not assume that the decomposition has small treewidth. Since our algorithm describes the polytope of correlated strategies directly, we get equilibrium finding in correlated strategies for free---instead of, say, having to run a double oracle algorithm. We show via experiments on a standard suite of games that our algorithm achieves state-of-the-art performance on all benchmark game classes except one. We also present, to our knowledge, the first experiments for this setting where both teams have more than one member.

Cutting plane methods play a significant role in modern solvers for tackling mixed-integer programming (MIP) problems. Proper selection of cuts would remove infeasible solutions in the early stage, thus largely reducing the computational burden without hurting the solution accuracy. However, the major cut selection approaches heavily rely on heuristics, which strongly depend on the specific problem at hand and thus limit their generalization capability. In this paper, we propose a data-driven and generalizable cut selection approach, named Cut Ranking, in the settings of multiple instance learning. To measure the quality of the candidate cuts, a scoring function, which takes the instance-specific cut features as inputs, is trained and applied in cut ranking and selection. In order to evaluate our method, we conduct extensive experiments on both synthetic datasets and real-world datasets. Compared with commonly used heuristics for cut selection, the learning-based policy has shown to be more effective, and is capable of generalizing over multiple problems with different properties. Cut Ranking has been deployed in an industrial solver for large-scale MIPs. In the online A/B testing of the product planning problems with more than $10^7$ variables and constraints daily, Cut Ranking has achieved the average speedup ratio of 12.42% over the production solver without any accuracy loss of solution.

Most existing results about last-iterate convergence of learning dynamics are limited to two-player zero-sum games, and only apply under rigid assumptions about what dynamics the players follow. In this paper we provide new results and techniques that apply to broader families of games and learning dynamics. First, we show that in a class of games that includes constant-sum polymatrix and strategically zero-sum games, the trajectories of dynamics such as optimistic mirror descent (OMD) exhibit a boundedness property, which holds even when players employ different algorithms and prediction mechanisms. This property enables us to obtain $O(1/\sqrt{T})$ rates and optimal $O(1)$ regret bounds. Our analysis also reveals a surprising property: OMD either reaches arbitrarily close to a Nash equilibrium or it outperforms the robust price of anarchy in efficiency. Moreover, for potential games we establish convergence to an $\epsilon$-equilibrium after $O(1/\epsilon^2)$ iterations for mirror descent under a broad class of regularizers, as well as optimal $O(1)$ regret bounds for OMD variants. Our framework also extends to near-potential games, and unifies known analyses for distributed learning in Fisher’s market model. Finally, we analyze the convergence, efficiency, and robustness of optimistic gradient descent (OGD) in general-sum continuous games.

In this paper we establish efficient and \emph{uncoupled} learning dynamics so that, when employed by all players in a general-sum multiplayer game, the \emph{swap regret} of each player after $T$ repetitions of the game is bounded by $O(\log T)$, improving over the prior best bounds of $O(\log^4 (T))$. At the same time, we guarantee optimal $O(\sqrt{T})$ swap regret in the adversarial regime as well. To obtain these results, our primary contribution is to show that when all players follow our dynamics with a \emph{time-invariant} learning rate, the \emph{second-order path lengths} of the dynamics up to time $T$ are bounded by $O(\log T)$, a fundamental property which could have further implications beyond near-optimally bounding the (swap) regret. Our proposed learning dynamics combine in a novel way \emph{optimistic} regularized learning with the use of \emph{self-concordant barriers}. Further, our analysis is remarkably simple, bypassing the cumbersome framework of higher-order smoothness recently developed by Daskalakis, Fishelson, and Golowich (NeurIPS'21).

We show that, for any sufficiently small fixed $\epsilon > 0$, when both players in a general-sum two-player (bimatrix) game employ optimistic mirror descent (OMD) with smooth regularization, learning rate $\eta = O(\epsilon^2)$ and $T = \Omega(poly(1/\epsilon))$ repetitions, either the dynamics reach an $\epsilon$-approximate Nash equilibrium (NE), or the average correlated distribution of play is an $\Omega(poly(\epsilon))$-strong coarse correlated equilibrium (CCE): any possible unilateral deviation does not only leave the player worse, but will decrease its utility by $\Omega(poly(\epsilon))$. As an immediate consequence, when the iterates of OMD are bounded away from being Nash equilibria in a bimatrix game, we guarantee convergence to an \emph{exact} CCE after only $O(1)$ iterations. Our results reveal that uncoupled no-regret learning algorithms can converge to CCE in general-sum games remarkably faster than to NE in, for example, zero-sum games. To establish this, we show that when OMD does not reach arbitrarily close to a NE, the (cumulative) regret of both players is not only negative, but decays linearly with time. Given that regret is the canonical measure of performance in online learning, our results suggest that cycling behavior of no-regret learning algorithms in games can be justified in terms of efficiency.

A shrinking market is a ubiquitous challenge faced by various industries. In this paper we formulate the first formal model of shrinking markets in multi-item settings, and study how mechanism design and machine learning can help preserve revenue in an uncertain, shrinking market. Via a sample-based learning mechanism, we prove the first guarantees on how much revenue can be preserved by truthful multi-item, multi-bidder auctions (for limited supply) when only a random unknown fraction of the population participates in the market. We first present a general reduction that converts any sufficiently rich auction class into a randomized auction robust to market shrinkage. Our main technique is a novel combinatorial construction called a winner diagram that concisely represents all possible executions of an auction on an uncertain set of bidders. Via a probabilistic analysis of winner diagrams, we derive a general possibility result: a sufficiently rich class of auctions always contains an auction that is robust to market shrinkage and market uncertainty. Our result has applications to important practically-constrained settings such as auctions with a limited number of winners. We then show how to efficiently learn an auction that is robust to market shrinkage by leveraging practically-efficient routines for solving the winner determination problem.

The incorporation of cutting planes within the branch-and-bound algorithm, known as branch-and-cut, forms the backbone of modern integer programming solvers. These solvers are the foremost method for solving discrete optimization problems and thus have a vast array of applications in machine learning, operations research, and many other fields. Choosing cutting planes effectively is a major research topic in the theory and practice of integer programming. We conduct a novel structural analysis of branch-and-cut that pins down how every step of the algorithm is affected by changes in the parameters defining the cutting planes added to the input integer program. Our main application of this analysis is to derive sample complexity guarantees for using machine learning to determine which cutting planes to apply during branch-and-cut. These guarantees apply to infinite families of cutting planes, such as the family of Gomory mixed integer cuts, which are responsible for the main breakthrough speedups of integer programming solvers. We exploit geometric and combinatorial structure of branch-and-cut in our analysis, which provides a key missing piece for the recent generalization theory of branch-and-cut.

A recent line of work has established uncoupled learning dynamics such that, when employed by all players in a game, each player's regret after $T$ repetitions grows polylogarithmically in $T$, an exponential improvement over the traditional guarantees within the no-regret framework. However, so far these results have only been limited to certain classes of games with structured strategy spaces---such as normal-form and extensive-form games. The question as to whether $O(\mathrm{polylog} T)$ regret bounds can be obtained for general convex and compact strategy sets---as is the case in many fundamental models in economics and multiagent systems---while retaining efficient strategy updates is an important question. In this paper, we answer this in the positive by establishing the first uncoupled learning algorithm with $O(\log T)$ per-player regret in general convex games, that is, games with concave utility functions supported on arbitrary convex and compact strategy sets. Our learning dynamics are based on an instantiation of optimistic follow-the-regularized-leader over an appropriately lifted space using a self-concordant regularizer that is peculiarly not a barrier for the feasible region. Our learning dynamics are efficiently implementable given access to a proximal oracle for the convex strategy set, leading to $O(\log\log T)$ per-iteration complexity; we also give extensions when access to only a linear optimization oracle is assumed. Finally, we adapt our dynamics to guarantee $O(\sqrt{T})$ regret in the adversarial regime. Even in those special cases where prior results apply, our algorithm improves over the state-of-the-art regret bounds either in terms of the dependence on the number of iterations or on the dimension of the strategy sets.

In adversarial team games, a team of players sequentially faces a team of adversaries. These games are the simplest setting with multiple players where cooperation and competition coexist, and it is known that the information asymmetry among the team members makes equilibrium approximation computationally hard. Although much effort has been spent designing scalable algorithms, the problem of solving large game instances is open. In this paper, we extend the successful approach of solving huge two-player zero-sum games, where a blueprint strategy is computed offline by using an abstract version of the game and then it is refined online, that is, during a playthrough. In particular, to the best of our knowledge, our paper provides the first method for online strategy refinement via subgame solving in adversarial team games. Our method, based on the team belief DAG, generates a gadget game and then refine the blueprint strategy by using column-generation approaches in anytime fashion. If the blueprint is sparse, then our whole algorithm runs end-to-end in polynomial time given a best-response oracle; in particular, it avoids expanding the whole team belief DAG, which has exponential worst-case size. We apply our method to a standard test suite, and we empirically show the performance improvement of the strategies thanks to subgame solving.

For common notions of correlated equilibrium in extensive-form games, computing an optimal (e.g., welfare-maximizing) equilibrium is NP-hard. Other equilibrium notions---communication and certification equilibria---augment the game with a mediator that has the power to both send and receive messages to and from the players---and, in particular, to remember the messages. In this paper, we investigate both notions in extensive-form games from a computational lens. We show that optimal equilibria in both notions can be computed in polynomial time, the latter under a natural additional assumption known in the literature. Our proof works by constructing a {\em mediator-augmented game} of polynomial size that explicitly represents the mediator's decisions and actions. Our framework allows us to define an entire family of equilibria by varying the mediator's information partition, the players' ability to lie, and the players' ability to deviate. From this perspective, we show that other notions of equilibrium, such as extensive-form correlated equilibrium, correspond to the mediator having imperfect recall. This shows that, at least among all these equilibrium notions, the hardness of computation is driven by the mediator's imperfect recall. As special cases of our general construction, we recover the polynomial-time algorithm of Conitzer & Sandholm [2004] for automated mechanism design in Bayes-Nash equilibria, and the correlation DAG algorithm of Zhang et al [2022] for optimal correlation. Our algorithm is especially scalable when the equilibrium notion is what we define as the full-certification equilibrium, where players cannot lie about their information but they can be silent. We back up our theoretical claims with experiments on a suite of standard benchmark games.

A recent emerging trend in the literature on learning in games has been concerned with providing faster learning dynamics for correlated and coarse correlated equilibria in normal-form games. Much less is known about the significantly more challenging setting of extensive-form games, which can capture both sequential and simultaneous moves, as well as imperfect information. In this paper we establish faster no-regret learning dynamics forextensive-form correlated equilibria (EFCE) in multiplayer general-sum imperfect-information extensive-form games. When all players follow our accelerated dynamics, the correlated distribution of play is an O(T-3/4)-approximate EFCE, where the O(·) notation suppresses parameters polynomial in the description of the game. This significantly improves over the best prior rate of O(T-1/2 ). To achieve this, we develop a framework for performing accelerated Phi-regret minimization via predictions. One of our key technical contributions---that enables us to employ our generic template---is to characterize the stability of fixed points associated with trigger deviation functions through a refined perturbation analysis of a structured Markov chain. Furthermore, for the simpler solution concept of extensive-form coarse correlated equilibrium (EFCCE) we give a new succinct closed-form characterization of the associated fixed points, bypassing the expensive computation of stationary distributions required for EFCE. Our results place EFCCE closer to normal-form coarse correlated equilibria in terms of the per-iteration complexity, although the former prescribes a much more compelling notion of correlation. Finally, experiments conducted on standard benchmarks corroborate our theoretical findings.

We study the problem of finding optimal correlated equilibria of various sorts in extensive-form games: normal-form coarse correlated equilibrium (NFCCE), extensive-form coarse correlated equilibrium (EFCCE), and extensive-form correlated equilibrium (EFCE). We make two primary contributions. First, we introduce a new algorithm for computing optimal equilibria in all three notions. Its runtime depends exponentially only on a parameter related to the information structure of the game. We also prove a fundamental complexity gap between NFCCE and the other two concepts. Second, we propose a two-sided column generation approach for use when the runtime or memory usage of the previous algorithm is prohibitive. Our algorithm improves upon an earlier one-sided approach by means of a new decomposition of correlated strategies which allows players to reoptimize their sequence-form strategies with respect to correlation plans which were previously added to the support. Experiments show that our techniques outperform the prior state of the art for computing optimal general-sum correlated equilibria. Funding: This work was supported by National Institutes of Health [Grant A240108S001]; Vannevar Bush Faculty Fellowship [Grant ONR N00014-23-1-2876]; National Science Foundation [Grants CCF-173355, IIS-190140, RI-1901403, RI-2312342]; and Army Research Office [Grants W911NF2010081, W911NF2210266].

Recently, Daskalakis, Fishelson, and Golowich (DFG) (NeurIPS ‘21) showed that if all agents in a multi-player general-sum normal-form game employ Optimistic Multiplicative Weights Update (OMWU), the external regret of every player is O(polylog(T)) after T repetitions of the game. In this paper we extend their result from external regret to internal and swap regret, thereby establishing uncoupled learning dynamics that converge to an approximate correlated equilibrium at the rate of O( T−1 ). This substantially improves over the prior best rate of convergence of O(T−3/4) due to Chen and Peng (NeurIPS ‘20), and it is optimal up to polylogarithmic factors.

Portfolio-based algorithm selection has seen tremendous practical success over the past two decades. This algorithm configuration procedure works by first selecting a portfolio of diverse algorithm parameter settings, and then, on a given problem instance, using an algorithm selector to choose a parameter setting from the portfolio with strong predicted performance. Oftentimes, both the portfolio and the algorithm selector are chosen using a training set of typical problem instances from the application domain at hand. In this paper, we provide the first provable guarantees for portfolio-based algorithm selection. We analyze how large the training set should be to ensure that the resulting algorithm selector's average performance over the training set is close to its future (expected) performance. This involves analyzing three key reasons why these two quantities may diverge: 1) the learning-theoretic complexity of the algorithm selector, 2) the size of the portfolio, and 3) the learning-theoretic complexity of the algorithm's performance as a function of its parameters. We introduce an end-to-end learning-theoretic analysis of the portfolio construction and algorithm selection together. We prove that if the portfolio is large, overfitting is inevitable, even with an extremely simple algorithm selector. With experiments, we illustrate a tradeoff exposed by our theoretical analysis: as we increase the portfolio size, we can hope to include a well-suited parameter setting for every possible problem instance, but it becomes impossible to avoid overfitting.

Blackwell approachability is a framework for reasoning about repeated games with vector-valued payoffs. We introduce predictive Blackwell approachability, where an estimate of the next payoff vector is given, and the decision maker tries to achieve better performance based on the accuracy of that estimator. In order to derive algorithms that achieve predictive Blackwell approachability, we start by showing a powerful connection between four well-known algorithms. Follow-the-regularized-leader (FTRL) and online mirror descent (OMD) are the most prevalent regret minimizers in online convex optimization. In spite of this prevalence, the regret matching (RM) and regret matching+ (RM+) algorithms have been preferred in the practice of solving large-scale games (as the local regret minimizers within the counterfactual regret minimization framework). We show that RM and RM+ are the algorithms that result from running FTRL and OMD, respectively, to select the halfspace to force at all times in the underlying Blackwell approachability game. By applying the predictive variants of FTRL or OMD to this connection, we obtain predictive Blackwell approachability algorithms, as well as predictive variants of RM and RM+. In experiments across 18 common zero-sum extensive-form benchmark games, we show that predictive RM+ coupled with counterfactual regret minimization converges vastly faster than the fastest prior algorithms (CFR+, DCFR, LCFR) across all games but two of the poker games, sometimes by two or more orders of magnitude.

Regret minimization has proved to be a versatile tool for tree-form sequential decision making and extensive-form games. In large two-player zero-sum imperfect-information games, modern extensions of counterfactual regret minimization (CFR) are currently the practical state of the art for computing a Nash equilibrium. Most regret-minimization algorithms for tree-form sequential decision making, including CFR, require (i) an exact model of the player’s decision nodes, observation nodes, and how they are linked, and (ii) full knowledge, at all times t, about the payoffs—even in parts of the decision space that are not encountered at time t. Recently, there has been growing interest towards relaxing some of those restrictions and making regret minimization applicable to settings for which reinforcement learning methods have traditionally been used—for example, those in which only black-box access to the environment is available. We give the first, to our knowledge, regret-minimization algorithm that guarantees sublinear regret with high probability even when requirement (i)—and thus also (ii)—is dropped. We formalize an online learning setting in which the strategy space is not known to the agent and gets revealed incrementally whenever the agent encounters new decision points. We give an efficient algorithm that achieves O(T^3/4) regret with high probability for that setting, even when the agent faces an adversarial environment. Our experiments show it significantly outperforms the prior algorithms for the problem, which do not have such guarantees. It can be used in any application for which regret minimization is useful: approximating Nash equilibrium or quantal response equilibrium, approximating coarse correlated equilibrium in multi-player games, learning a best response, learning safe opponent exploitation, and online play against an unknown opponent/environment.

Tree-form sequential decision making (TFSDM) extends classical one-shot decision making by modeling tree-form interactions between an agent and a potentially adversarial environment. It captures the online decision-making problems that each player faces in an extensive-form game, as well as Markov decision processes and partially-observable Markov decision processes where the agent conditions on observed history. Over the past decade, there has been considerable effort into designing online optimization methods for TFSDM. Virtually all of that work has been in the full-feedback setting, where the agent has access to counterfactuals, that is, information on what would have happened had the agent chosen a different action at any decision node. Little is known about the bandit setting, where that assumption is reversed (no counterfactual information is available), despite this latter setting being well understood for almost 20 years in one-shot decision making. In this paper, we give the first algorithm for the bandit linear optimization problem for TFSDM that offers both (i) linear-time iterations (in the size of the decision tree) and (ii) O(sqrt(T)) cumulative regret in expectation compared to any fixed strategy, at all times T. This is made possible by new results that we derive, which may have independent uses as well: 1) geometry of the dilated entropy regularizer, 2) autocorrelation matrix of the natural sampling scheme for sequence-form strategies, 3) construction of an unbiased estimator for linear losses for sequence-form strategies, and 4) a refined regret analysis for mirror descent when using the dilated entropy regularizer.

Often---for example in war games, strategy video games, and financial simulations---the game is given to us only as a black-box simulator in which we can play it. In these settings, since the game may have unknown nature action distributions (from which we can only obtain samples) and/or be too large to expand fully, it can be difficult to compute strategies with guarantees on exploitability. Recent work (Zhang and Sandholm 2020) resulted in a notion of certificate for extensive-form games that allows exploitability guarantees while not expanding the full game tree. However, that work assumed that the black box could sample or expand arbitrary nodes of the game tree at any time, and that a series of exact game solves (via, for example, linear programming) can be conducted to compute the certificate. Each of those two assumptions severely restricts the practical applicability of that method. In this work, we relax both of the assumptions. We show that high-probability certificates can be obtained with a black box that can do nothing more than play through games, using only a regret minimizer as a subroutine. As a bonus, we obtain an equilibrium-finding algorithm with $\tilde O(1/\sqrt{T})$ convergence rate in the extensive-form game setting that does not rely on a sampling strategy with lower-bounded reach probabilities (which MCCFR assumes). We demonstrate experimentally that, in the black-box setting, our methods are able to provide nontrivial exploitability guarantees while expanding only a small fraction of the game tree.

We focus on the problem of finding an optimal strategy for a team of players that faces an opponent in an imperfect-information zero-sum extensive-form game. Team members are not allowed to communicate during play but can coordinate before the game. In this setting, it is known that the best the team can do is sample a profile of potentially randomized strategies (one per player) from a joint (a.k.a. correlated) probability distribution at the beginning of the game. In this paper, we first provide new modeling results about computing such an optimal distribution by drawing a connection to a different literature on extensive-form correlation. Second, we provide an algorithm that allows one for capping the number of profiles employed in the solution. This begets an anytime algorithm by increasing the cap. We find that often a handful of well-chosen such profiles suffices to reach optimal utility for the team. This enables team members to reach coordination through a simple and understandable plan. Finally, inspired by this observation and leveraging theoretical concepts that we introduce, we develop an efficient column-generation algorithm for finding an optimal distribution for the team. We evaluate it on a suite of common benchmark games. It is three orders of magnitude faster than the prior state of the art on games that the latter can solve and it can also solve several games that were previously unsolvable.

We develop a new framework for designing truthful, high-revenue (combinatorial) auctions for limited supply. Our mechanism learns within an instance. It generalizes and improves over previously-studied random-sampling mechanisms. It first samples a participatory group of bidders, then samples several learning groups of bidders from the remaining pool of bidders, learns a high-revenue auction from the learning groups, and finally runs that auction on the participatory group. Previous work on random-sampling mechanisms focused primarily on unlimited supply. Limited supply poses additional significant technical challenges, since allocations of items to bidders must be feasible. We prove guarantees on the performance of our mechanism based on a market-shrinkage term and a new complexity measure we coin partition discrepancy. Partition discrepancy simultaneously measures the intrinsic complexity of the mechanism class and the uniformity of the set of bidders. We then introduce new auction classes that can be parameterized in a way that does not depend on the number of bidders participating, and prove strong guarantees for these classes. We show how our mechanism can be implemented efficiently by leveraging practically-efficient routines for solving winner determination. Finally, we show how to use structural revenue maximization to decide what auction class to use with our framework when there is a constraint on the number of learning groups.

Cutting-plane methods have enabled remarkable successes in integer programming over the last few decades. State-of-the-art solvers integrate a myriad of cutting-plane techniques to speed up the underlying tree-search algorithm used to find optimal solutions. In this paper we provide sample complexity bounds for cut-selection in branch-and-cut (B&C). Given a training set of integer programs sampled from an application-specific input distribution and a family of cut selection policies, these guarantees bound the number of samples sufficient to ensure that using any policy in the family, the size of the tree B&C builds on average over the training set is close to the expected size of the tree B&C builds. We first bound the sample complexity of learning cutting planes from the canonical family of Chvátal-Gomory cuts. Our bounds handle any number of waves of any number of cuts and are fine tuned to the magnitudes of the constraint coefficients. Next, we prove sample complexity bounds for more sophisticated cut selection policies that use a combination of scoring rules to choose from a family of cuts. Finally, beyond the realm of cutting planes for integer programming, we develop a general abstraction of tree search that captures key components such as node selection and variable selection. For this abstraction, we bound the sample complexity of learning a good policy for building the search tree.

In two-player zero-sum extensive-form games, Nash equilibrium prescribes optimal strategies against perfectly rational opponents. However, it does not guarantee rational play in parts of the game tree that can only be reached by the players making mistakes. This can be problematic when operationalizing equilibria in the real world among imperfect players. Trembling-hand refinements are a sound remedy to this issue, and are subsets of Nash equilibria that are designed to handle the possibility that any of the players may make mistakes. In this paper, we initiate the study of equilibrium refinements for settings where one of the players is perfectly rational (the ``machine'') and the other may make mistakes. As we show, this endeavor has many pitfalls: many intuitively appealing approaches to refinement fail in various ways. On the positive side, we introduce a modification of the classical quasi-perfect equilibrium (QPE) refinement, which we call the one-sided quasi-perfect equilibrium. Unlike QPE, one-sided QPE only accounts for mistakes from one player and assumes that no mistakes will be made by the machine. We present experiments on standard benchmark games and an endgame from the famous man-machine match where the AI Libratus was the first to beat top human specialist professionals in heads-up no-limit Texas hold'em poker. We show that one-sided QPE can be computed more efficiently than all known prior refinements, paving the way to wider adoption of Nash equilibrium refinements in settings with perfectly rational machines (or humans perfectly actuating machine-generated strategies) that interact with players prone to mistakes. We also show that one-sided QPE tends to play better than a Nash equilibrium strategy against imperfect opponents.

In imperfect-information games, subgame solving is significantly more challenging than in perfect-information games, but in the last few years, such techniques have been developed. They were the key ingredient to the milestone of superhuman play in no-limit Texas hold'em poker. Current subgame-solving techniques analyze the entire common-knowledge closure of the player's current information set, that is, the smallest set of nodes within which it is common knowledge that the current node lies. While this is acceptable in games like poker where the common-knowledge closure is relatively small, many practical games have more complex information structure, which renders the common-knowledge closure impractically large to enumerate or even reasonably approximate. We introduce an approach that overcomes this obstacle, by instead working with only low-order knowledge. Our approach allows an agent, upon arriving at an infoset, to basically prune any node that is no longer reachable, thereby massively reducing the game tree size relative to the common-knowledge subgame. We prove that, as is, our approach can increase exploitability compared to the blueprint strategy. However, we develop three avenues by which safety can be guaranteed. First, safety is guaranteed if the results of subgame solves are incorporated back into the blueprint. Second, we provide a method where safety is achieved by limiting the infosets at which subgame solving is performed. Third, we prove that our approach, when applied at every infoset reached during play, achieves a weaker notion of equilibrium, which we coin affine equilibrium, and which may be of independent interest. We show that affine equilibria cannot be exploited by any Nash strategy of the opponent, so an opponent who wishes to exploit must open herself to counter-exploitation. Even without the safety-guaranteeing additions, experiments on medium-sized games show that our approach always reduced exploitability in practical games even when applied at every infoset, and a depth-limited version of it led to---to our knowledge---the first strong AI for the challenge problem dark chess.

We study the application of iterative first-order methods to the problem of computing equilibria of large-scale two-player extensive-form games. First-order methods must typically be instantiated with a regularizer that serves as a distance-generating function for the decision sets of the players. For the case of two-player zero-sum games, the state-of-the-art theoretical convergence rate for Nash equilibrium is achieved by using the dilated entropy function. In this paper, we introduce a new entropy-based distance-generating function for two-player zero-sum games, and show that this function achieves significantly better strong convexity properties than the dilated entropy, while maintaining the same easily-implemented closed-form proximal mapping. Extensive numerical simulations show that these superior theoretical properties translate into better numerical performance as well.

Algorithms often have tunable parameters that impact performance metrics such as runtime and solution quality. For many algorithms used in practice, no parameter settings admit meaningful worst-case bounds, so the parameters are made available for the user to tune. Alternatively, parameters may be tuned implicitly within the proof of a worst-case guarantee. Worst-case instances, however, may be rare or nonexistent in practice. A growing body of research has demonstrated that data-driven algorithm design can lead to significant improvements in performance. This approach uses a training set of problem instances sampled from an unknown, application-specific distribution and returns a parameter setting with strong average performance on the training set.

The design of efficient combination therapies is a difficult key challenge in the treatment of complex diseases such as cancers. The large heterogeneity of cancers and the large number of available drugs renders exhaustive in vivo or even in vitro investigation of possible treatments impractical. In recent years, sophisticated mechanistic, ordinary differential equation-based pathways models that can predict treatment responses at a molecular level have been developed. However, surprisingly little effort has been put into leveraging these models to find novel therapies. In this paper we use for the first time, to our knowledge, a large-scale state-of-the-art pan-cancer signaling pathway model to identify candidates for novel combination therapies to treat individual cancer cell lines from various tissues (e.g., minimizing proliferation while keeping dosage low to avoid adverse side effects) and populations of heterogeneous cancer cell lines (e.g., minimizing the maximum or average proliferation across the cell lines while keeping dosage low). We also show how our method can be used to optimize the drug combinations used in sequential treatment plans-that is, optimized sequences of potentially different drug combinations-providing additional benefits. In order to solve the treatment optimization problems, we combine the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) algorithm with a significantly more scalable sampling scheme for truncated Gaussian distributions, based on a Hamiltonian Monte-Carlo method. These optimization techniques are independent of the signaling pathway model, and can thus be adapted to find treatment candidates for other complex diseases than cancers as well, as long as a suitable predictive model is available.

Algorithms typically come with tunable parameters that have a considerable impact on the computational resources they consume. Too often, practitioners must hand-tune the parameters, a tedious and error-prone task. A recent line of research provides algorithms that return nearly-optimal parameters from within a finite set. These algorithms can be used when the parameter space is infinite by providing as input a random sample of parameters. This data-independent discretization, however, might miss pockets of nearly-optimal parameters: prior research has presented scenarios where the only viable parameters lie within an arbitrarily small region. We provide an algorithm that learns a finite set of promising parameters from within an infinite set. Our algorithm can help compile a configuration portfolio, or it can be used to select the input to a configuration algorithm for finite parameter spaces. Our approach applies to any configuration problem that satisfies a simple yet ubiquitous structure: the algorithm's performance is a piecewise constant function of its parameters. Prior research has exhibited this structure in domains from integer programming to clustering.

Coarse correlation models strategic interactions of rational agents complemented by a correlation device which is a mediator that can recommend behavior but not enforce it. Despite being a classical concept in the theory of normal-form games since 1978, not much is known about the merits of coarse correlation in extensive-form settings. In this paper, we consider two instantiations of the idea of coarse correlation in extensive-form games: normal-form coarse-correlated equilibrium (NFCCE), already defined in the literature, and extensive-form coarse-correlated equilibrium (EFCCE), a new solution concept that we introduce. We show that EFCCEs are a subset of NFCCEs and a superset of the related extensive-form correlated equilibria. We also show that, in n-player extensive-form games, social-welfare-maximizing EFCCEs and NFCCEs are bilinear saddle points, and give new efficient algorithms for the special case of two-player games with no chance moves. Experimentally, our proposed algorithm for NFCCE is two to four orders of magnitude faster than the prior state of the art.

Automating algorithm configuration is growing increasingly necessary as algorithms come with more and more tunable parameters. It is common to tune parameters using machine learning, optimizing algorithmic performance (runtime or solution quality, for example) using a training set of problem instances from the specific domain at hand. We investigate a fundamental question about these techniques: how large should the training set be to ensure that a parameter’s average empirical performance over the training set is close to its expected, future performance? We answer this question for algorithm configuration problems that exhibit a widely-applicable structure: the algorithm’s performance as a function of its parameters can be approximated by a “simple” function. We show that if this approximation holds under the L$\infty$-norm, we can provide strong sample complexity bounds, but if the approximation holds only under the Lp-norm for p < $\infty$, it is not possible to provide meaningful sample complexity bounds in the worst case. We empirically evaluate our bounds in the context of integer programming, obtaining sample complexity bounds that are up to 700 times smaller than the previously best-known bounds.

Monte-Carlo counterfactual regret minimization (MCCFR) is the state-of-the-art algorithm for solving sequential games that are too large for full tree traversals. It works by using gradient estimates that can be computed via sampling. However, stochastic methods for sequential games have not been investigated extensively beyond MCCFR. In this paper we develop a new framework for developing stochastic regret minimization methods. This framework allows us to use any regret-minimization algorithm, coupled with any gradient estimator. The MCCFR algorithm can be analyzed as a special case of our framework, and this analysis leads to significantly stronger theoretical guarantees on convergence, while simultaneously yielding a simplified proof. Our framework allows us to instantiate several new stochastic methods for solving sequential games. We show extensive experiments on five games, where some variants of our methods outperform MCCFR.

Computational equilibrium finding in large zero-sum extensive-form imperfect-information games has led to significant recent AI breakthroughs. The fastest algorithms for the problem are new forms of counterfactual regret minimization (Brown & Sandholm, 2019). In this paper we present a totally different approach to the problem, which is competitive and often orders of magnitude better than the prior state of the art. The equilibrium-finding problem can be formulated as a linear program (LP) (Koller et al., 1994), but solving it as an LP has not been scalable due to the memory requirements of LP solvers, which can often be quadratically worse than CFR-based algorithms. We give an efficient practical algorithm that factors a large payoff matrix into a product of two matrices that are typically dramatically sparser. This allows us to express the equilibrium-finding problem as a linear program with size only a logarithmic factor worse than CFR, and thus allows linear program solvers to run on such games. With experiments on poker endgames, we demonstrate in practice, for the first time, that modern linear program solvers are competitive against even game-specific modern variants of CFR in solving large extensive-form games, and can be used to compute exact solutions unlike iterative algorithms like CFR.

A two-part tariff is a pricing scheme that consists of an up-front lump sum fee and a per unit fee. Various products in the real world are sold via a menu, or list, of two-part tariffs---for example gym memberships, cell phone data plans, etc. We study learning high-revenue menus of two-part tariffs from buyer valuation data, in the setting where the mechanism designer has access to samples from the distribution over buyers' values rather than an explicit description thereof. Our algorithms have clear direct uses, and provide the missing piece for the recent generalization theory of two-part tariffs. We present a polynomial time algorithm for optimizing one two-part tariff. We also present an algorithm for optimizing a length-L menu of two-part tariffs with run time exponential in L but polynomial in all other problem parameters. We then generalize the problem to multiple markets. We prove how many samples suffice to guarantee that a two-part tariff scheme that is feasible on the samples is also feasible on a new problem instance with high probability. We then show that computing revenue-maximizing feasible prices is hard even for buyers with additive valuations. Then, for buyers with identical valuation distributions, we present a condition that is sufficient for the two-part tariff scheme from the unsegmented setting to be optimal for the market-segmented setting. Finally, we prove a generalization result that states how many samples suffice so that we can compute the unsegmented solution on the samples and still be guaranteed that we get a near-optimal solution for the market-segmented setting with high probability.

Unlike normal-form games, where correlated equilibria have been studied for more than 45 years, extensive-form correlation is still generally not well understood. Part of the reason for this gap is that the sequential nature of extensive-form games allows for a richness of behaviors and incentives that are not possible in normal-form settings. This richness translates to a significantly different complexity landscape surrounding extensive-form correlated equilibria. As of today, it is known that finding an optimal extensive-form correlated equilibrium (EFCE), extensive-form coarse correlated equilibrium (EFCCE), or normal-form coarse correlated equilibrium (NFCCE) in a two-player extensive-form game is computationally tractable when the game does not include chance moves, and intractable when the game involves chance moves. In this paper we significantly refine this complexity threshold by showing that, in two-player games, an optimal correlated equilibrium can be computed in polynomial time, provided that a certain condition is satisfied. We show that the condition holds, for example, when all chance moves are public, that is, both players observe all chance moves. This implies that an optimal EFCE, EFCCE and NFCCE can be computed in polynomial time in the game size in two-player games with public chance moves.

In barter exchanges, participants swap goods with one another without exchanging money; these exchanges are often facilitated by a central clearinghouse, with the goal of maximizing the aggregate quality (or number) of swaps. Barter exchanges are subject to many forms of uncertainty--in participant preferences, the feasibility and quality of various swaps, and so on. Our work is motivated by kidney exchange, a real-world barter market in which patients in need of a kidney transplant swap their willing living donors, in order to find a better match. Modern exchanges include 2- and 3-way swaps, making the kidney exchange clearing problem NP-hard. Planned transplants often \emph{fail} for a variety of reasons--if the donor organ is rejected by the recipient's medical team, or if the donor and recipient are found to be medically incompatible. Due to 2- and 3-way swaps, failed transplants can ``cascade'' through an exchange; one US-based exchange estimated that about $85\%$ of planned transplants failed in 2019. Many optimization-based approaches have been designed to avoid these failures; however most exchanges cannot implement these methods, due to legal and policy constraints. Instead, we consider a setting where exchanges can \emph{query} the preferences of certain donors and recipients--asking whether they would accept a particular transplant. We characterize this as a two-stage decision problem, in which the exchange program (a) queries a small number of transplants before committing to a matching, and (b) constructs a matching according to fixed policy. We show that selecting these edges is a challenging combinatorial problem, which is non-monotonic and non-submodular, in addition to being NP-hard. We propose both a greedy heuristic and a Monte Carlo tree search, which outperforms previous approaches, using experiments on both synthetic data and real kidney exchange data from the United Network for Organ Sharing.

In many game settings, the game is not explicitly given but is only accessible by playing it. While there have been impressive demonstrations in such settings, prior techniques have not offered safety guarantees, that is, guarantees on the game-theoretic exploitability of the computed strategies. In this paper we introduce an approach that shows that it is possible to provide exploitability guarantees in such settings without ever exploring the entire game. We introduce a notion of a certificatae of an extensive-form approximate Nash equilibrium. For verifying a certificate, we give an algorithm that runs in time linear in the size of the certificate rather than the size of the whole game. In zero-sum games, we further show that an optimal certificate---given the exploration so far---can be computed with any standard game-solving algorithm (e.g., using a linear program or counterfactual regret minimization). However, unlike in the cases of normal form or perfect information, we show that certain families of extensive-form games do not have small approximate certificates, even after making extremely nice assumptions on the structure of the game. Despite this difficulty, we find experimentally that very small certificates, even exact ones, often exist in large and even in infinite games. Overall, our approach enables one to try one's favorite exploration strategies while offering exploitability guarantees, thereby decoupling the exploration strategy from the equilibrium-finding process.

Counterfactual regret minimization (CFR) is a family of iterative algorithms that are the most popular and, in practice, fastest approach to approximately solving large imperfectinformation games. In this paper we introduce novel CFR variants that 1) discount regrets from earlier iterations in various ways (in some cases differently for positive and negative regrets), 2) reweight iterations in various ways to obtain the output strategies, 3) use a non-standard regret minimizer and/or 4) leverage “optimistic regret matching”. They lead to dramatically improved performance in many settings. For one, we introduce a variant that outperforms CFR+, the prior state-of-the-art algorithm, in every game tested, including large-scale realistic settings. CFR+ is a formidable benchmark: no other algorithm has been able to outperform it. Finally, we show that, unlike CFR+, many of the important new variants are compatible with modern imperfect-informationgame pruning techniques and one is also compatible with sampling in the game tree.

Regret minimization is a powerful tool for solving large-scale extensive-form games. State-of-the-art methods rely on minimizing regret locally at each decision point. In this work we derive a new framework for regret minimization on sequential decision problems and extensive-form games with general compact convex sets at each decision point and general convex losses, as opposed to prior work which has been for simplex decision points and linear losses. We call our framework laminar regret decomposition. It generalizes the CFR algorithm to this more general setting. Furthermore, our framework enables a new proof of CFR even in the known setting, which is derived from a perspective of decomposing polytope regret, thereby leading to an arguably simpler interpretation of the algorithm. Our generalization to convex compact sets and convex losses allows us to develop new algorithms for several problems: regularized sequential decision making, regularized Nash equilibria in zero-sum extensive-form games, and computing approximate extensive-form perfect equilibria. Our generalization also leads to the first regret-minimization algorithm for computing reduced-normal-form quantal response equilibria based on minimizing local regrets. Experiments show that our framework leads to algorithms that scale at a rate comparable to the fastest variants of counterfactual regret minimization for computing Nash equilibrium, and therefore our approach leads to the first algorithm for computing quantal response equilibria in extremely large games. Our algorithms for (quadratically) regularized equilibrium finding are orders of magnitude faster than the fastest algorithms for Nash equilibrium finding; this suggests regret-minimization algorithms based on decreasing regularization for Nash equilibrium finding as future work. Finally we show that our framework enables a new kind of scalable opponent exploitation approach.

Equilibrium refinements are important in extensive-form (i.e., tree-form) games, where they amend weaknesses of the Nash equilibrium concept by requiring sequential rationality and other beneficial properties. One of the most attractive refinement concepts is quasi-perfect equilibrium. While quasiperfection has been studied in extensive-form games, it is poorly understood in Stackelberg settings—that is, settings where a leader can commit to a strategy—which are important for modeling, for example, security games. In this paper, we introduce the axiomatic definition of quasi-perfect Stackelberg equilibrium. We develop a broad class of game perturbation schemes that lead to them in the limit. Our class of perturbation schemes strictly generalizes prior perturbation schemes introduced for the computation of (non-Stackelberg) quasi-perfect equilibria. Based on our perturbation schemes, we develop a branch-and-bound algorithm for computing a quasi-perfect Stackelberg equilibrium. It leverages a perturbed variant of the linear program for computing a Stackelberg extensive-form correlated equilibrium. Experiments show that our algorithm can be used to find an approximate quasi-perfect Stackelberg equilibrium in games with thousands of nodes.

Counterfactual Regret Minimization (CFR) is the leading algorithm for solving large imperfect-information games. It converges to an equilibrium by iteratively traversing the game tree. In order to deal with extremely large games, abstraction is typically applied before running CFR. The abstracted game is solved with tabular CFR, and its solution is mapped back to the full game. This process can be problematic because aspects of abstraction are often manual and domain specific, abstraction algorithms may miss important strategic nuances of the game, and there is a chicken-and-egg problem because determining a good abstraction requires knowledge of the equilibrium of the game. This paper introduces Deep Counterfactual Regret Minimization, a form of CFR that obviates the need for abstraction by instead using deep neural networks to approximate the behavior of CFR in the full game. We show that Deep CFR is principled and achieves strong performance in large poker games. This is the first non-tabular variant of CFR to be successful in large games.

The CFR framework has been a powerful tool for solving large-scale extensive-form games in practice. However, the theoretical rate at which past CFR-based algorithms converge to the Nash equilibrium is on the order of $O(T^{-1/2})$, where $T$ is the number of iterations. In contrast, first-order methods can be used to achieve a $O(T^{-1})$ dependence on iterations, yet these methods have been less successful in practice. In this work we present the first CFR variant that breaks the square-root dependence on iterations. By combining and extending recent advances on predictive and stable regret minimizers for the matrix-game setting we show that it is possible to leverage “optimistic” regret minimizers to achieve a $O(T^{-3/4})$ convergence rate within CFR. This is achieved by introducing a new notion of stable-predictivity, and by setting the stability of each counterfactual regret minimizer relative to its location in the decision tree. Experiments show that this method is faster than the original CFR algorithm, although not as fast as newer variants, in spite of their worst-case $O(T^{-1/2})$ dependence on iterations.

Regret minimization is a powerful tool for solving large-scale problems; it was recently used in breakthrough results for large-scale extensive-form game solving. This was achieved by composing simplex regret minimizers into an overall regret-minimization framework for extensive-form game strategy spaces. In this paper we study the general composability of regret minimizers. We derive a calculus for constructing regret minimizers for composite convex sets that are obtained from convexity-preserving operations on simpler convex sets. We show that local regret minimizers for the simpler sets can be combined with additional regret minimizers into an aggregate regret minimizer for the composite set. As one application, we show that the CFR framework can be constructed easily from our framework. We also show ways to include curtailing (constraining) operations into our framework. For one, they enable the construction of CFR generalization for extensive-form games with general convex strategy constraints that can cut across decision points.

We study the performance of optimistic regret-minimization algorithms for both minimizing regret in, and computing Nash equilibria of, zero-sum extensive-form games. In order to apply these algorithms to extensive-form games, a distance-generating function is needed. We study the use of the dilated entropy and dilated Euclidean distance functions. For the dilated Euclidean distance function we prove the first explicit bounds on the strong-convexity parameter for general treeplexes. Furthermore, we show that the use of dilated distance-generating functions enable us to decompose the mirror descent algorithm, and its optimistic variant, into local mirror descent algorithms at each information set. This decomposition mirrors the structure of the counterfactual regret minimization framework, and enables important techniques in practice, such as distributed updates and pruning of cold parts of the game tree. Our algorithms provably converge at a rate of $T^{-1}$, which is superior to prior counterfactual regret minimization algorithms. We experimentally compare to the popular algorithm CFR+, which has a theoretical convergence rate of $T^{-0.5}$ in theory, but is known to often converge at a rate of $T^{-1}$, or better, in practice. We give an example matrix game where CFR+ experimentally converges at a relatively slow rate of $T^{-0.74}$, whereas our optimistic methods converge faster than $T^{-1}$. We go on to show that our fast rate also holds in the Kuhn poker game, which is an extensive-form game. For games with deeper game trees however, we find that CFR+ is still faster. Finally we show that when the goal is minimizing regret, rather than computing a Nash equilibrium, our optimistic methods can outperform CFR+, even in deep game trees.

Self-play methods based on regret minimization have become the state of the art for computing Nash equilibria in large two-players zero-sum extensive-form games. These methods fundamentally rely on the hierarchical structure of the players' sequential strategy spaces to construct a regret minimizer that recursively minimizes regret at each decision point in the game tree. In this paper, we introduce the first efficient regret minimization algorithm for computing extensive-form correlated equilibria in large two-player general-sum games with no chance moves. Designing such an algorithm is significantly more challenging than designing one for the Nash equilibrium counterpart, as the constraints that define the space of correlation plans lack the hierarchical structure and might even form cycles. We show that some of the constraints are redundant and can be excluded from consideration, and present an efficient algorithm that generates the space of extensive-form correlation plans incrementally from the remaining constraints. This structural decomposition is achieved via a special convexity-preserving operation that we coin scaled extension. We show that a regret minimizer can be designed for a scaled extension of any two convex sets, and that from the decomposition we then obtain a global regret minimizer. Our algorithm produces feasible iterates. Experiments show that it significantly outperforms prior approaches and for larger problems it is the only viable option.

While Nash equilibrium in extensive-form games is well understood, very little is known about the properties of extensive-form correlated equilibrium (EFCE), both from a behavioral and from a computational point of view. In this setting, the strategic behavior of players is complemented by an external device that privately recommends moves to agents as the game progresses; players are free to deviate at any time, but will then not receive future recommendations. Our contributions are threefold. First, we show that an EFCE can be formulated as the solution to a bilinear saddle-point problem. To showcase how this novel formulation can inspire new algorithms to compute EFCEs, we propose a simple subgradient descent method which exploits this formulation and structural properties of EFCEs. Our method has better scalability than the prior approach based on linear programming. Second, we propose two benchmark games, which we hope will serve as the basis for future evaluation of EFCE solvers. These games were chosen so as to cover two natural application domains for EFCE: conflict resolution via a mediator, and bargaining and negotiation. Third, we document the qualitative behavior of EFCE in our proposed games. We show that the social-welfare-maximizing equilibria in these games are highly nontrivial and exhibit surprisingly subtle sequential behavior that so far has not received attention in the literature.

Stackelberg equilibria have become increasingly important as a solution concept in computational game theory, largely inspired by practical problems such as security settings. In practice, however, there is typically uncertainty regarding the model about the opponent. This paper is, to our knowledge, the first to investigate Stackelberg equilibria under uncertainty in extensive-form games, one of the broadest classes of game. We introduce robust Stackelberg equilibria, where the uncertainty is about the opponent’s payoffs, as well as ones where the opponent has limited lookahead and the uncertainty is about the opponent’s node evaluation function. We develop a new mixed-integer program for the deterministic limited-lookahead setting. We then extend the program to the robust setting for Stackelberg equilibrium under unlimited and under limited lookahead by the opponent. We show that for the specific case of interval uncertainty about the opponent’s payoffs (or about the opponent’s node evaluations in the case of limited lookahead), robust Stackelberg equilibria can be computed with a mixed-integer program that is of the same asymptotic size as that for the deterministic setting.

Tree search algorithms, such as branch-and-bound, are the most widely used tools for solving combinatorial problems. These algorithms recursively partition the search space to find an optimal solution. To keep the tree small, it is crucial to carefully decide, when expanding a tree node, which variable to branch on at that node to partition the remaining space. Many partitioning techniques have been proposed, but no theory describes which is optimal. We show how to use machine learning to determine an optimal weighting of any set of partitioning procedures for the instance distribution at hand using samples. Via theory and experiments, we show that learning to branch is both practical and hugely beneficial.

We initiate the study of equilibrium refinements based on trembling-hand perfection in extensive-form games with commitment strategies, that is, where one player commits to a strategy first. We show that the standard strong (and weak) Stackelberg equilibria are not suitable for trembling-hand perfection, because the limit of a sequence of such strong (weak) Stackelberg commitment strategies of a perturbed game may not be a strong (weak) Stackelberg equilibrium itself. However, we show that the universal set of all Stackelberg equilibria (i.e., those that are optimal for at least some follower response function) is natural for trembling- hand perfection: it does not suffer from the problem above. We also prove that determining the existence of a Stackelberg equilibrium--refined or not--that gives the leader expected value at least v is NP-hard. This significantly extends prior complexity results that were specific to strong Stackelberg equilibrium.

A fundamental challenge in imperfect-information games is that states do not have well-defined values. As a result, depth-limited search algorithms used in single-agent settings and perfect-information games do not apply. This paper introduces a principled way to conduct depth-limited solving in imperfect-information games by allowing the opponent to choose among a number of strategies for the remainder of the game at the depth limit. Each one of these strategies results in a different set of values for leaf nodes. This forces an agent to be robust to the different strategies an opponent may employ. We demonstrate the effectiveness of this approach by building a master-level heads-up no-limit Texas hold'em poker AI that defeats two prior top agents using only a 4-core CPU and 16 GB of memory. Developing such a powerful agent would have previously required a supercomputer.

Nash equilibrium strategies have the known weakness that they do not prescribe rational play in situations that are reached with zero probability according to the strategies themselves, for example, if players have made mistakes. Trembling-hand refinements---such as extensive-form perfect equilibria and quasi-perfect equilibria---remedy this problem in sound ways. Despite their appeal, they have not received attention in practice since no known algorithm for computing them scales beyond toy instances. In this paper, we design an exact polynomial-time algorithm for finding trembling-hand equilibria in zero-sum extensive-form games. It is several orders of magnitude faster than the best prior ones, numerically stable, and quickly solves game instances with tens of thousands of nodes in the game tree. It enables, for the first time, the use of trembling-hand refinements in practice.

Recent milestones in equilibrium computation, such as the success of Libratus, show that it is possible to compute strong solutions to two-player zero-sum games in theory and practice. This is not the case for games with more than two players, which remain one of the main open challenges in computational game theory. This paper focuses on zero-sum games where a team of players faces an opponent, as is the case, for example, in Bridge, collusion in poker, and many non-recreational applications such as war, where the colluders do not have time or means of communicating during battle, collusion in bidding, where communication during the auction is illegal, and coordinated swindling in public. The possibility for the team members to communicate before game play—that is, coordinate their strategies ex ante—makes the use of behavioral strategies unsatisfactory. The reasons for this are closely related to the fact that the team can be represented as a single player with imperfect recall. We propose a new game representation, the realization form, that generalizes the sequence form but can also be applied to imperfect-recall games. Then, we use it to derive an auxiliary game that is equivalent to the original one. It provides a sound way to map the problem of finding an optimal ex-ante-correlated strategy for the team to the well-understood Nash equilibrium-finding problem in a (larger) two-player zero-sum perfect-recall game. By reasoning over the auxiliary game, we devise an anytime algorithm, fictitious team-play, that is guaranteed to converge to an optimal coordinated strategy for the team against an optimal opponent, and that is dramatically faster than the prior state-of-the-art algorithm for this problem.

There has been tremendous recent progress on equilibrium-finding algorithms for zero-sum imperfect-information extensive-form games, but there has been a puzzling gap between theory and practice. First-order methods have significantly better theoretical convergence rates than any counterfactual-regret minimization (CFR) variant. Despite this, CFR variants have been favored in practice. Experiments with first-order methods have only been conducted on small- and medium-sized games because those methods are complicated to implement in this setting, and because CFR variants have been enhanced extensively for over a decade they perform well in practice. In this paper we show that a particular first-order method, a state-of-the-art variant of the excessive gap technique---instantiated with the dilated entropy distance function---can efficiently solve large real-world problems competitively with CFR and its variants. We show this on large endgames encountered by the Libratus poker AI, which recently beat top human poker specialist professionals at no-limit Texas hold'em. We show experimental results on our variant of the excessive gap technique as well as a prior version. We introduce a numerically friendly implementation of the smoothed best response computation associated with first-order methods for extensive-form game solving. We present, to our knowledge, the first GPU implementation of a first-order method for extensive-form games. We present comparisons of several excessive gap technique and CFR variants.

Abstraction has long been a key component in the practical solving of large-scale extensive-form games. Despite this, abstraction remains poorly understood. There have been some recent theoretical results but they have been confined to specific assumptions on abstraction structure and are specific to various disjoint types of abstraction, and specific solution concepts, for example, exact Nash equilibria or strategies with bounded immediate regret. In this paper we present a unified framework for analyzing abstractions that can express all types of abstractions and solution concepts used in prior papers with performance guarantees---while maintaining comparable bounds on abstraction quality. Moreover, our framework gives an exact decomposition of abstraction error in a much broader class of games, albeit only in an ex-post sense, as our results depend on the specific strategy chosen. Nonetheless, we use this ex-post decomposition along with slightly weaker assumptions than prior work to derive generalizations of prior bounds on abstraction quality. We also show, via counterexample, that such assumptions are necessary for some games. Finally, we prove the first bounds for how $\epsilon$-Nash equilibria computed in abstractions perform in the original game. This is important because often one cannot afford to compute an exact Nash equilibrium in the abstraction. All our results apply to general-sum n-player games.

One of the most tantalizing and long-standing open problems in mechanism design is profit maximization in multi-item, multi-buyer settings. Much of the literature surrounding this problem rests on the strong assumption that the mechanism designer knows the distribution over buyers' values. In reality, this information is rarely available. The support of the distribution alone is often doubly exponential, so obtaining and storing the distribution is impractical. We relax this assumption and instead assume that the mechanism designer only has a set of samples from the distribution. We develop learning-theoretic foundations of sample-based mechanism design. In particular, we provide generalization guarantees which bound the difference between the empirical profit of a mechanism over a set of samples and its expected profit on the unknown distribution.

We present a complete algorithm for finding an epsilon-Nash equilibrium, for arbitrarily small epsilon, in games with more than two players. The best prior complete algorithm has significantly worse complexity and has, to our knowledge, never been implemented. The main components of our tree-search-based method are a node-selection strategy, an exclusion oracle, and a subdivision scheme. The node-selection strategy determines the next region to be explored---based on the region's size and an estimate of whether the region contains an equilibrium. The exclusion oracle provides a provably correct sufficient condition for there not to exist an equilibrium in the region. The subdivision scheme determines how the region is split if it cannot be excluded. Unlike well-known incomplete methods, our method does not need to proceed locally, which avoids it getting stuck in a local minimum that may be far from any actual equilibrium. The run time grows rapidly with the game size, and this suggests a hybrid scheme where one of the relatively fast prior incomplete algorithms is run, and if it fails to find an equilibrium, then our method is used.

Regret minimization is widely used in determining strategies for imperfect-information games and in online learning. In large games, computing the regrets associated with a single iteration can be slow. For this reason, pruning — in which parts of the decision tree are not traversed in every iteration — has emerged as an essential method for speeding up iterations in large games. The ability to prune is a primary reason why the Counterfactual Regret Minimization (CFR) algorithm using regret matching has emerged as the most popular iterative algorithm for imperfect-information games, despite its relatively poor convergence bound. In this paper, we introduce dynamic thresholding, in which a threshold is set at every iteration such that any action in the decision tree with probability below the threshold is set to zero probability. This enables pruning for the first time in a wide range of algorithms. We prove that dynamic thresholding can be applied to Hedge while increasing its convergence bound by only a constant factor in terms of number of iterations. Experiments demonstrate a substantial improvement in performance for Hedge as well as the excessive gap technique.

Kidney exchanges are organized markets where patients swap willing but incompatible donors. In the last decade, kidney exchanges grew from small and regional to large and national — and soon, international. This growth results in more lives saved, but exacerbates the empirical hardness of the NP-complete problem of optimally matching patients to donors. State-of-the-art matching engines use integer programming techniques to clear fielded kidney exchanges, but these methods must be tailored to specific models and objective functions, and may fail to scale to larger exchanges. In this paper, we observe that if the kidney exchange compatibility graph can be encoded by a constant number of patient and donor attributes, the clearing problem is solvable in polynomial time. We give necessary and sufficient conditions for losslessly shrinking the representation of an arbitrary compatibility graph. Then, using real compatibility graphs from the UNOS US-wide kidney exchange, we show how many attributes are needed to encode real graphs. The experiments show that, indeed, small numbers of attributes suffice.

Iterative algorithms such as Counterfactual Regret Minimization (CFR) are the most popular way to solve large zero-sum imperfect-information games. In this paper we introduce Best-Response Pruning (BRP), an improvement to iterative algorithms such as CFR that allows poorly-performing actions to be temporarily pruned. We prove that when using CFR in zero-sum games, adding BRP will asymptotically prune any action that is not part of a best response to some Nash equilibrium. This leads to provably faster convergence and lower space requirements. Experiments show that BRP results in a factor of 7 reduction in space, and the reduction factor increases with game size.

No-regret learning has emerged as a powerful tool for solving extensive-form games. This was facilitated by the counterfactual-regret minimization (CFR) framework, which relies on the instantiation of regret minimizers for simplexes at each information set of the game. We use an instantiation of the CFR framework to develop algorithms for solving behaviorally-constrained (and, as a special case, perturbed in the Selten sense) extensive-form games, which allows us to compute approximate Nash equilibrium refinements. Nash equilibrium refinements are motivated by a major deficiency in Nash equilibrium: it provides virtually no guarantees on how it will play in parts of the game tree that are reached with zero probability. Refinements can mend this issue, but have not been adopted in practice, mostly due to a lack of scalable algorithms. We show that, compared to standard algorithms, our method finds solutions that have substantially better refinement properties, while enjoying a convergence rate that is comparable to that of state-of-the-art algorithms for Nash equilibrium computation both in theory and practice.

No-limit Texas Hold'em is the most popular variant of poker in the world. Heads-up no-limit Texas Hold'em is the main benchmark challenge for AI in imperfect-information games. We present Libratus, the first - and so far only - AI to defeat top human professionals in that game. Libratus's architecture features three main modules, each of which has new algorithms: pre-computing a solution to an abstraction of the game which provides a high-level blueprint for the strategy of the AI, a new nested subgame-solving algorithm which repeatedly calculates a more detailed strategy as play progresses, and a self-improving module which augments the pre-computed blueprint over time.

A kidney exchange is a centrally-administered barter market where patients swap their willing yet incompatible donors. Modern kidney exchanges use 2-cycles, 3-cycles, and chains initiated by non-directed donors (altruists who are willing to give a kidney to anyone) as the means for swapping. We propose significant generalizations to kidney exchange. We allow more than one donor to donate in exchange for their desired patient receiving a kidney. We also allow for the possibility of a donor willing to donate if any of a number of patients receive kidneys. Furthermore, we combine these notions and generalize them. The generalization is to exchange among organ clubs, where a club is willing to donate organs outside the club if and only if the club receives organs from outside the club according to given specifications. We prove that unlike in the standard model, the uncapped clearing problem is NP-complete. We also present the notion of operation frames that can be used to sequence the operations across batches, and present integer programming formulations for the market clearing problems for these new types of organ exchanges. Experiments show that in the single-donation setting, operation frames improve planning by 34% - 51%. Allowing up to two donors to donate in exchange for one kidney donated to their designated patient yields a further increase in social welfare.

Nash equilibrium is a popular solution concept for solving imperfect-information games in practice. However, it has a major drawback: it does not preclude suboptimal play in branches of the game tree that are not reached in equilibrium. Equilibrium refinements can mend this issue, but have experienced little practical adoption. This is largely due to a lack of scalable algorithms.Sparse iterative methods, in particular first-order methods, are known to be among the most effective algorithms for computing Nash equilibria in large-scale two-player zero-sum extensive-form games. In this paper, we provide, to our knowledge, the first extension of these methods to equilibrium refinements. We develop a smoothing approach for behavioral perturbations of the convex polytope that encompasses the strategy spaces of players in an extensive-form game. This enables one to compute an approximate variant of extensive-form perfect equilibria. Experiments show that our smoothing approach leads to solutions with dramatically stronger strategies at information sets that are reached with low probability in approximate Nash equilibria, while retaining the overall convergence rate associated with fast algorithms for Nash equilibrium. This has benefits both in approximate equilibrium finding (such approximation is necessary in practice in large games) where some probabilities are low while possibly heading toward zero in the limit, and exact equilibrium computation where the low probabilities are actually zero.

Poker has been a challenge problem in AI and game theory for decades. As a game of imperfect information it involves obstacles not present in games like chess and Go, and requires totally different techniques. No program had been able to beat top players in large poker games. Until now! In January 2017, our AI, Libratus, beat a team of four top specialist professionals in heads-up no-limit Texas hold'em, which has 10^161 decision points. This game is the main benchmark challenge for imperfect-information game solving. Libratus is the only AI that has beat top humans at this game. Libratus is powered by new algorithms in each of its three main modules: 1. computing blueprint (approximate Nash equilibrium) strategies before the event, 2. novel nested endgame solving during play, and 3. fixing its own strategy to play even closer to equilibrium based on what holes the opponents have been able to identify and exploit. These domain-independent algorithms have potential applicability to a variety of real-world imperfect-information games such as negotiation, business strategy, cybersecurity, physical security, military applications, strategic pricing, product portfolio planning, certain areas of finance, auctions, political campaigns, and steering biological adaptation and evolution, for example, for medical treatment planning.

In imperfect-information games, the optimal strategy in a subgame may depend on the strategy in other, unreached subgames. Thus a subgame cannot be solved in isolation and must instead consider the strategy for the entire game as a whole, unlike perfect-information games. Nevertheless, it is possible to first approximate a solution for the whole game and then improve it in individual subgames. This is referred to as subgame solving. We introduce subgame-solving techniques that outperform prior methods both in theory and practice. We also show how to adapt them, and past subgame-solving techniques, to respond to opponent actions that are outside the original action abstraction; this significantly outperforms the prior state-of-the-art approach, action translation. Finally, we show that subgame solving can be repeated as the game progresses down the game tree, leading to far lower exploitability. These techniques were a key component of Libratus, the first AI to defeat top humans in heads-up no-limit Texas hold'em poker.

Sparse iterative methods, in particular first-order methods, are known to be among the most effective in solving large-scale two-player zero-sum extensive-form games. The convergence rates of these methods depend heavily on the properties of the distance-generating function that they are based on. We investigate the acceleration of first-order methods for solving extensive-form games through better design of the dilated entropy function---a class of distance-generating functions related to the domains associated with the extensive-form games. By introducing a new weighting scheme for the dilated entropy function, we develop the first distance-generating function for the strategy spaces of sequential games that only a logarithmic dependence on the branching factor of the player. This result improves the convergence rate of several first-order methods by a factor of Ω(bdd), where b is the branching factor of the player, and d is the depth of the game tree.

Kidney exchange, where candidates with organ failure trade incompatible but willing donors, is a life-saving alternative to the deceased donor waitlist, which has inadequate supply to meet demand. While fielded kidney exchanges see huge benefit from altruistic kidney donors (who give an organ without a paired needy candidate), a significantly higher medical risk to the donor deters similar altruism with livers. In this paper, we begin by exploring the idea of large-scale liver exchange, and show on demographically accurate data that vetted kidney exchange algorithms can be adapted to clear such an exchange at the nationwide level. We then propose cross-organ donation where kidneys and livers can be bartered for each other. We show theoretically that this multi-organ exchange provides linearly more transplants than running separate kidney and liver exchanges. This linear gain is a product of altruistic kidney donors creating chains that thread through the liver pool; it exists even when only a small but constant portion of the donors on the kidney side of the pool are willing to donate a liver lobe. We support this result experimentally on demographically accurate multi-organ exchanges. We conclude with thoughts regarding the fielding of a nationwide liver or joint liver-kidney exchange from a legal and computational point of view.

Counterfactual Regret Minimization (CFR) is a popular iterative algorithm for approximating Nash equilibria in imperfect-information multi-step two-player zero-sum games. We introduce the first general, principled method for warm starting CFR. Our approach requires only a strategy for each player, and accomplishes the warm start at the cost of a single traversal of the game tree. The method provably warm starts CFR to as many iterations as it would have taken to reach a strategy profile of the same quality as the input strategies, and does not alter the convergence bounds of the algorithms. Unlike prior approaches to warm starting, ours can be applied in all cases. Our method is agnostic to the origins of the input strategies. For example, they can be based on human domain knowledge, the observed strategy of a strong agent, the solution of a coarser abstraction, or the output of some algorithm that converges rapidly at first but slowly as it gets closer to an equilibrium. Experiments demonstrate that one can improve overall convergence in a game by first running CFR on a smaller, coarser abstraction of the game and then using the strategy in the abstract game to warm start CFR in the full game.

Kidney exchange is a type of barter market where patients exchange willing but incompatible donors. These exchanges are conducted via cycles---where each incompatible patient-donor pair in the cycle both gives and receives a kidney---and chains, which are started by an altruist donor who does not need a kidney in return. Finding the best combination of cycles and chains is hard. The leading algorithms for this optimization problem use either branch and price — a combination of branch and bound and column generation — or constraint generation. We show a correctness error in the leading prior branch-and-price-based approach [Glorie et al. 2014]. We develop a provably correct fix to it, which also necessarily changes the algorithm's complexity, as well as other improvements to the search algorithm. Next, we compare our solver to the leading constraint-generation-based solver and to the best prior correct branch-and-price-based solver. We focus on the setting where chains have a length cap. A cap is desirable in practice since if even one edge in the chain fails, the rest of the chain fails: the cap precludes very long chains that are extremely unlikely to execute and instead causes the solution to have more parallel chains and cycles that are more likely to succeed. We work with the UNOS nationwide kidney exchange, which uses a chain cap. Algorithms from our group autonomously make the transplant plans for that exchange. On that real data and demographically-accurate generated data, our new solver scales significantly better than the prior leading approaches.

The design of revenue-maximizing combinatorial auctions, i.e. multi item auctions over bundles of goods, is one of the most fundamental problems in computational economics, unsolved even for two bidders and two items for sale. In the traditional economic models, it is assumed that the bidders' valuations are drawn from an underlying distribution and that the auction designer has perfect knowledge of this distribution. Despite this strong and oftentimes unrealistic assumption, it is remarkable that the revenue-maximizing combinatorial auction remains unknown. In recent years, automated mechanism design has emerged as one of the most practical and promising approaches to designing high-revenue combinatorial auctions. The most scalable automated mechanism design algorithms take as input samples from the bidders' valuation distribution and then search for a high-revenue auction in a rich auction class. In this work, we provide the first sample complexity analysis for the standard hierarchy of deterministic combinatorial auction classes used in automated mechanism design. In particular, we provide tight sample complexity bounds on the number of samples needed to guarantee that the empirical revenue of the designed mechanism on the samples is close to its expected revenue on the underlying, unknown distribution over bidder valuations, for each of the auction classes in the hierarchy. In addition to helping set automated mechanism design on firm foundations, our results also push the boundaries of learning theory. In particular, the hypothesis functions used in our contexts are defined through multi stage combinatorial optimization procedures, rather than simple decision boundaries, as are common in machine learning.

A kidney exchange is an organized barter market where patients in need of a kidney swap willing but incompatible donors. Determining an optimal set of exchanges is theoretically and empirically hard. Traditionally, exchanges took place in cycles, with each participating patient-donor pair both giving and receiving a kidney. The recent introduction of chains, where a donor without a paired patient triggers a sequence of donations without requiring a kidney in return, increased the efficacy of fielded kidney exchanges---while also dramatically raising the empirical computational hardness of clearing the market in practice. While chains can be quite long, unbounded-length chains are not desirable: planned donations can fail before transplant for a variety of reasons, and the failure of a single donation causes the rest of that chain to fail, so parallel shorter chains are better in practice.

Imperfect-recall abstraction has emerged as the leading paradigm for practical large-scale equilibrium computation in imperfect-information games. However, imperfect-recall abstractions are poorly understood, and only weak algorithm-specific guarantees on solution quality are known. We develop the first general, algorithm-agnostic, solution quality guarantees for Nash equilibria and approximate self-trembling equilibria computed in imperfect-recall abstractions, when implemented in the original (perfect-recall) game. Our results are for a class of games that generalizes the only previously known class of imperfect-recall abstractions for which any such results have been obtained. Further, our analysis is tighter in two ways, each of which can lead to an exponential reduction in the solution quality error bound.

The leading approach for solving large imperfect-information games is automated abstraction followed by running an equilibrium-finding algorithm. We introduce a distributed version of the most commonly used equilibrium-finding algorithm, counterfactual regret minimization (CFR), which enables CFR to scale to dramatically larger abstractions and numbers of cores. The new algorithm begets constraints on the abstraction so as to make the pieces running on different computers disjoint. We introduce an algorithm for generating such abstractions while capitalizing on state-of-the-art abstraction ideas such as imperfect recall and the earth-mover's-distance similarity metric. Our techniques enabled an equilibrium computation of unprecedented size on a supercomputer with a high inter-blade memory latency. Prior approaches run slowly on this architecture. Our approach also leads to a significant improvement over using the prior best approach on a large shared-memory server with low memory latency. Finally, we introduce a family of post-processing techniques that outperform prior ones. We applied these techniques to generate an agent for two-player no-limit Texas Hold'em. It won the 2014 Annual Computer Poker Competition, beating each opponent with statistical significance.

The preferred treatment for kidney failure is a transplant; however, demand for donor kidneys far outstrips supply. Kidney exchange, an innovation where willing but incompatible patient-donor pairs can exchange organs- — via barter cycles and altruist-initiated chains —provides a life-saving alternative.Typically, fielded exchanges act myopically, considering only the current pool of pairs when planning the cycles and chains. Yet kidney exchange is inherently dynamic, with participants arriving and departing. Also, many planned exchange transplants do not go to surgery due to various failures. So, it is important to consider the future when matching. Motivated by our experience running the computational side of a large nationwide kidney exchange, we present FutureMatch, a framework for learning to match in a general dynamic model. FutureMatch takes as input a high-level objective (e.g., "maximize graft survival of transplants over time'') decided on by experts, then automatically (i) learns based on data how to make this objective concrete and (ii) learns the ``means'' to accomplish this goal — a task, in our experience, that humans handle poorly. It uses data from all live kidney transplants in the US since 1987 to learn the quality of each possible match; it then learns the potentials of elements of the current input graph offline (e.g., potentials of pairs based on features such as donor and patient blood types), translates these to weights, and performs a computationally feasible batch matching that incorporates dynamic, failure-aware considerations through the weights. We validate FutureMatch on real fielded exchange data. It results in higher values of the objective. Furthermore, even under economically inefficient objectives that enforce equity, it yields better solutions for the efficient objective (which does not incorporate equity) than traditional myopic matching that uses the efficiency objective.

We present a credit-based matching mechanism for dynamic barter markets — and kidney exchange in particular — that is both strategy proof and efficient, that is, it guarantees truthful disclosure of donor-patient pairs from the transplant centers and results in the maximum global matching. Furthermore, the mechanism is individually rational in the sense that, in the long run, it guarantees each transplant center more matches than the center could have achieved alone. The mechanism does not require assumptions about the underlying distribution of compatibility graphs — a nuance that has previously produced conflicting results in other aspects of theoretical kidney exchange. Our results apply not only to matching via 2-cycles: the matchings can also include cycles of any length and altruist-initiated chains, which is important at least in kidney exchanges. The mechanism can also be adjusted to guarantee immediate individual rationality at the expense of economic efficiency, while preserving strategy proofness via the credits. This circumvents a well-known impossibility result in static kidney exchange concerning the existence of an individually rational, strategy-proof, and maximal mechanism. We show empirically that the mechanism results in significant gains on data from a national kidney exchange that includes 59% of all US transplant centers.

Living organisms adapt to challenges through evolution. This has proven to be a key difficulty in developing therapies, since the organisms evolve resistance.I propose the wild idea of steering evolution strategically — using computational game theory for (typically incomplete-information) multistage games and opponent exploitation techniques. A sequential contingency plan for steering evolution is constructed computationally for the setting at hand. In the biological context, the opponent (e.g., a disease) has a systematic handicap because it evolves myopically. This can be exploited by computing trapping strategies that cause the opponent to evolve into states where it can be handled effectively. Potential application classes include therapeutics at the population, individual, and molecular levels (drug design), as well as cell repurposing and synthetic biology.

Most real-world games and many recreational games are games of incomplete information. Over the last dozen years, abstraction has emerged as a key enabler for solving large incomplete-information games. First, the game is abstracted to generate a smaller, abstract game that is strategically similar to the original game. Second, an approximate equilibrium is computed in the abstract game. Third, the strategy from the abstract game is mapped back to the original game. In this paper, I will review key developments in the field. I present reasons for abstracting games, and point out the issue of abstraction pathology. I then review the practical algorithms for information abstraction and action abstraction. I then cover recent theoretical breakthroughs that beget bounds on the quality of the strategy from the abstract game, when measured in the original game. I then discuss how to reverse map the opponent's action into the abstraction if the opponent makes a move that is not in the abstraction. Finally, I discuss other topics of current and future research.

Counterfactual Regret Minimization (CFR) is a leading algorithm for finding a Nash equilibrium in large zero-sum imperfect-information games. CFR is an iterative algorithm that repeatedly traverses the game tree, updating regrets at each information set.We introduce an improvement to CFR that prunes any path of play in the tree, and its descendants, that has negative regret. It revisits that sequence at the earliest subsequent CFR iteration where the regret could have become positive, had that path been explored on every iteration. The new algorithm maintains CFR's convergence guarantees while making iterations significantly faster---even if previously known pruning techniques are used in the comparison. This improvement carries over to CFR+, a recent variant of CFR. Experiments show an order of magnitude speed improvement, and the relative speed improvement increases with the size of the game.

The stochastic matching problem deals with finding a maximum matching in a graph whose edges are unknown but can be accessed via queries. This is a special case of stochastic k-set packing, where the problem is to find a maximum packing of sets, each of which exists with some probability. In this paper, we provide edge and set query algorithms for these two problems, respectively, that provably achieve some fraction of the omniscient optimal solution. Our main theoretical result for the stochastic matching (i.e., 2-set packing) problem is the design of an adaptive algorithm that queries only a constant number of edges per vertex and achieves a (1-ε) fraction of the omniscient optimal solution, for an arbitrarily small ε > 0. Moreover, this adaptive algorithm performs the queries in only a constant number of rounds. We complement this result with a non-adaptive (i.e., one round of queries) algorithm that achieves a (0.5 - ε) fraction of the omniscient optimum. We also extend both our results to stochastic k-set packing by designing an adaptive algorithm that achieves a (2/k - ε) fraction of the omniscient optimal solution, again with only O(1) queries per element. This guarantee is close to the best known polynomial-time approximation ratio of 3/k+1 -ε for the deterministic k-set packing problem [Furer 2013].

We study the problem of computing a Nash equilibrium in large-scale two-player zero-sum extensive-form games. While this problem can be solved in polynomial time, first-order or regret-based methods are usually preferred for large games. Regret-based methods have largely been favored in practice, in spite of their theoretically inferior convergence rates. In this paper we investigate the acceleration of first-order methods both theoretically and experimentally. An important component of many first-order methods is a distance-generating function. Motivated by this, we investigate a specific distance-generating function, namely the dilated entropy function, over treeplexes, which are convex polytopes that encompass the strategy spaces of perfect-recall extensive-form games. We develop significantly stronger bounds on the associated strong convexity parameter. In terms of extensive-form game solving, this improves the convergence rate of several first-order methods by a factor of O((#information sets ⋅ depth ⋅ M)/(2depth)) where M is the maximum value of the l1 norm over the treeplex encoding the strategy spaces. Experimentally, we investigate the performance of three first-order methods (the excessive gap technique, mirror prox, and stochastic mirror prox) and compare their performance to the regret-based algorithms. In order to instantiate stochastic mirror prox, we develop a class of gradient sampling schemes for game trees. Equipped with our distance-generating function and sampling scheme, we find that mirror prox and the excessive gap technique outperform the prior regret-based methods for finding medium accuracy solutions

Regret matching is a widely-used algorithm for learning how to act. We begin by proving that regrets on actions in one setting (game) can be transferred to warm start the regrets for solving a different setting with same structure but different payoffs that can be written as a function of parameters. We prove how this can be done by carefully discounting the prior regrets. This provides, to our knowledge, the first principled warm-starting method for no-regret learning. It also extends to warm-starting the widely-adopted counterfactual regret minimization (CFR) algorithm for large incomplete-information games; we show this experimentally as well. We then study optimizing a parameter vector for a player in a two-player zero-sum game (e.g., optimizing bet sizes to use in poker). We propose a custom gradient descent algorithm that provably finds a locally optimal parameter vector while leveraging our warm-start theory to significantly save regret-matching iterations at each step. It optimizes the parameter vector while simultaneously finding an equilibrium. We present experiments in no-limit Leduc Hold'em and no-limit Texas Hold'em to optimize bet sizing. This amounts to the first action abstraction algorithm (algorithm for selecting a small number of discrete actions to use from a continuum of actions---a key preprocessing step for solving large games using current equilibrium-finding algorithms) with convergence guarantees for extensive-form games.

The fair division of indivisible goods has long been an important topic in economics and, more recently, computer science. We investigate the existence of envy-free allocations of indivisible goods, that is, allocations where each player values her own allocated set of goods at least as highly as any other player's allocated set of goods. Under additive valuations, we show that even when the number of goods is larger than the number of agents by a linear fraction, envy-free allocations are unlikely to exist. We then show that when the number of goods is larger by a logarithmic factor, such allocations exist with high probability. We support these results experimentally and show that the asymptotic behavior of the theory holds even when the number of goods and agents is quite small. We demonstrate that there is a sharp phase transition from nonexistence to existence of envy-free allocations, and that on average the computational problem is hardest at that transition.

Kidney exchange, where candidates with organ failure trade incompatible but willing donors, is a life-saving alternative to the deceased donor waitlist, which has inadequate supply to meet demand. While fielded kidney exchanges see huge benefit from altruistic kidney donors (who give an organ without a paired needy candidate), a significantly higher medical risk to the donor deters similar altruism with livers. In this paper, we begin by proposing the idea of liver exchange, and show on demographically accurate data that vetted kidney exchange algorithms can be adapted to clear such an exchange at the nationwide level. We then explore cross-organ donation where kidneys and livers can be bartered for each other. We show theoretically that this multi-organ exchange provides linearly more transplants than running separate kidney and liver exchanges; this linear gain is a product of altruistic kidney donors creating chains that thread through the liver pool. We support this result experimentally on demographically accurate multi-organ exchanges. We conclude with thoughts regarding the fielding of a nationwide liver or joint liver-kidney exchange from a legal and computational point of view.

There is often a large disparity between the size of a game we wish to solve and the size of the largest instances solvable by the best algorithms; for example, a popular variant of poker has about $10^{165}$ nodes in its game tree, while the currently best approximate equilibrium-finding algorithms scale to games with around $10^{12}$ nodes. In order to approximate equilibrium strategies in these games, the leading approach is to create a sufficiently small strategic approximation of the full game, called an abstraction, and to solve that smaller game instead. The leading abstraction algorithm for imperfect-information games generates abstractions that have imperfect recall and are distribution aware, using $k$-means with the earth mover's distance metric to cluster similar states together. A distribution-aware abstraction groups states together at a given round if their full distributions over future strength are similar (as opposed to, for example, just the expectation of their strength). The leading algorithm considers distributions over future strength at the final round of the game. However, one might benefit by considering the trajectory of distributions over strength in all future rounds, not just the final round. An abstraction algorithm that takes all future rounds into account is called potential aware. We present the first algorithm for computing potential-aware imperfect-recall abstractions using earth mover's distance. Experiments on no-limit Texas Hold'em show that our algorithm improves performance over the previously best approach.

Abstraction has emerged as a key component in solving extensive-form games of incomplete information. However, lossless abstractions are typically too large to solve, so lossy abstraction is needed. All prior lossy abstraction algorithms for extensive-form games either 1) had no bounds on solution quality or 2) depended on specific equilibrium computation approaches, limited forms of abstraction, and only decreased the number of information sets rather than nodes in the game tree. We introduce a theoretical framework that can be used to give bounds on solution quality for any perfect-recall extensive-form game. The framework uses a new notion for mapping abstract strategies to the original game, and it leverages a new equilibrium refinement for analysis. Using this framework, we develop the first general lossy extensive-form game abstraction method with bounds. Experiments show that it finds a lossless abstraction when one is available and lossy abstractions when smaller abstractions are desired. While our framework can be used for lossy abstraction, it is also a powerful tool for lossless abstraction if we set the bound to zero. Prior abstraction algorithms typically operate level by level in the game tree. We introduce the extensive-form game tree isomorphism and action subset selection problems, both important problems for computing abstractions on a level-by-level basis. We show that the former is graph isomorphism complete, and the latter NP-complete. We also prove that level-by-level abstraction can be too myopic and thus fail to find even obvious lossless abstractions.

A descending (multi-item) clock auction (DCA) is a mechanism for buying items from multiple potential sellers. In the DCA, bidder-specific prices are decremented over the course of the auction. In each round, each bidder might accept or decline his offer price. Accepting means the bidder is willing to sell at that price. Rejecting means the bidder will not sell at that price or a lower price. DCAs have been proposed as the method for procuring spectrum from existing holders in the FCC's imminent incentive auctions so spectrum can be repurposed to higher-value uses. However, the DCA design has lacked a way to determine the prices to offer the bidders in each round. This is a recognized, important, and timely problem.

The openness of wireless communication and the recent development of software-defined radio technology, respectively, provide a low barrier and a wide range of capabilities for misbehavior, attacks, and defenses against attacks. In this work we present finite-energy jamming games, a game model that allows a jammer and sender to choose (1) whether to transmit or sleep, (2) a power level to transmit with, and (3) what channel to transmit on. We also allow the jammer to choose on how many channels it simultaneously attacks. A major addition in finite-energy jamming games is that the jammer and sender both have a limited amount of energy which is drained according to the actions a player takes.

Strong Nash equilibrium (SNE) is an appealing solution concept when rational agents can form coalitions. A strategy profile is an SNE if no coalition of agents can benefit by deviating. We present the first general-purpose algorithms for SNE finding in games with more than two agents. An SNE must simultaneously be a Nash equilibrium (NE) and the optimal solution of multiple non-convex optimization problems. This makes even the derivation of necessary and sufficient mathematical equilibrium constraints difficult. We show that forcing an SNE to be resilient only to pure-strategy deviations by coalitions, unlike for NEs, is only a necessary condition here. Second, we show that the application of Karush-Kuhn-Tucker conditions leads to another set of necessary conditions that are not sufficient. Third, we show that forcing the Pareto efficiency of an SNE for each coalition with respect to coalition correlated strategies is sufficient but not necessary. We then develop a tree search algorithm for SNE finding. At each node, it calls an oracle to suggest a candidate SNE and then verifies the candidate. We show that our new necessary conditions can be leveraged to make the oracle more powerful. Experiments validate the overall approach and show that the new conditions significantly reduce search tree size compared to using NE conditions alone.

Most algorithmic matches in fielded kidney exchanges do not result in an actual transplant. In this paper, we address the problem of cycles and chains in a proposed match failing after the matching algorithm has committed to them. We show that failure-aware kidney exchange can significantly increase the expected number of lives saved (i) in theory, on random graph models; (ii) on real data from kidney exchange match runs between 2010 and 2012; (iii) on synthetic data generated via a model of dynamic kidney exchange. From the computational viewpoint, we design a branch-and-price-based optimal clearing algorithm specifically for the probabilistic exchange clearing problem and show that this new solver scales well on large simulated data, unlike prior clearing algorithms.

One typically proves infeasibility in satisfiability/constraint satisfaction (or optimality in integer programming) by constructing a tree certificate. However, deciding how to branch in the search tree is hard, and impacts search time drastically. We explore the power of a simple paradigm, that of throwing random darts into the assignment space and then using information gathered by that dart to guide what to do next. Such guidance is easy to incorporate into state-of-the-art solvers. This method seems to work well when the number of short certificates of infeasibility is moderate, suggesting the overhead of throwing darts can be countered by the information gained by these darts. We explore results supporting this suggestion both on instances from a new generator where the size and number of short certificates can be controlled, and on industral instances from the annual SAT competition.

In many dynamic matching applications — especially high-stakes ones — the competitive ratios of prior-free online algorithms are unacceptably poor. The algorithm should take distributional information about possible futures into account in deciding what action to take now. This is typically done by drawing sample trajectories of possible futures at each time period, but may require a prohibitively large number of trajectories or prohibitive memory and/or computation to decide what action to take. Instead, we propose to learn potentials of elements (e.g., vertices) of the current problem. Then, at run time, we simply run an offline matching algorithm at each time period, but subtracting out in the objective the potentials of the elements used up in the matching. We apply the approach to kidney exchange. Kidney exchanges enable willing but incompatible patient-donor pairs (vertices) to swap donors. These swaps typically include cycles longer than two pairs and chains triggered by altruistic donors. Fielded exchanges currently match myopically, maximizing the number of patients who get kidneys in an offline fashion at each time period. Myopic matching is sub-optimal; the clearing problem is dynamic since patients, donors, and altruists appear and expire over time. We theoretically compare the power of using potentials on increasingly large elements: vertices, edges, cycles, and the entire graph (optimum). Then, experiments show that by learning vertex potentials, our algorithm matches more patients than the current practice of clearing myopically. It scales to exchanges orders of magnitude beyond those handled by the prior dynamic algorithm.

Designing revenue-optimal auctions for various settings is perhaps the most important, yet sometimes most elusive, problem in mechanism design. Spiteful bidders have been intensely studied recently, especially because spite occurs in many applications in multiagent system and electronic commerce. We derive the optimal auction for such bidders (as well as bidders that are altruistic). It is a generalization of Myerson’s (1981) auction. It chooses an allocation that maximizes agents’ virtual valuations, but for a generalized definition of virtual valuation. The payment rule is less intuitive. For one, it takes each bidder’s own report into consideration when determining his payment. Moreover, bidders pay even if the seller keeps the item; a similar phenomenon has been shown in other settings with neg- ative externalities (Jehiel, Moldovanu, and Stacchetti 1996; Deng and Pekec 2011). On the other hand, a novel aspect of our auction is that it sometimes subsidizes losers when the item is sold to some other bidder. We also derive a revenue equivalence theorem for this setting. Using it, we generate a short proof of (a slight generalization of) the previously known result that, in two-bidder settings with independently uniformly drawn valuations, second-price auctions yield greater expected revenue than first-price auctions. Finally, we present a template for comparing the expected revenues of any two auction mechanisms that have the same allocation rule (for the valuations distributions at hand).

In proof-of-payment transit systems, passengers are legally required to purchase tickets before entering but are not physically forced to do so. Instead, patrol units move about the transit system, inspecting the tickets of passengers, who face fines if caught fare evading. The deterrence of such fines depends on the unpredictability and effectiveness of the patrols. In this paper, we present TRUSTS, an application for scheduling randomized patrols for fare inspection in transit systems. TRUSTS models the problem of computing patrol strategies as a leader-follower Stackelberg game where the objective is to deter fare evasion and hence maximize revenue. This problem differs from previously studied Stackelberg settings in that the leader strategies must satisfy massive temporal and spatial constraints; moreover, unlike in these counterterrorism-motivated Stackelberg applications, a large fraction of the ridership might realistically consider fare evasion, and so the number of followers is potentially huge. A third key novelty in our work is deliberate simplification of leader strategies to make patrols easier to be executed. We present an efficient algorithm for computing such patrol strategies and present experimental results using real-world ridership data from the Los Angeles Metro Rail system. The Los Angeles County Sheriff’s department has begun trials of TRUSTS.

We consider the problem of playing a finitely-repeated two-player zero-sum game safely---that is, guaranteeing at least the value of the game per period in expectation regardless of the strategy used by the opponent. Playing a stage-game equilibrium strategy at each time step clearly guarantees safety, and prior work has conjectured that it is impossible to simultaneously deviate from a stage-game equilibrium (in hope of exploiting a suboptimal opponent) and to guarantee safety. We show that such profitable deviations are indeed possible---specifically, in games where certain types of 'gift' strategies exist, which we define formally. We show that the set of strategies constituting such gifts can be strictly larger than the set of iteratively weakly-dominated strategies; this disproves another recent conjecture which states that all non-iteratively-weakly-dominated strategies are best responses to each equilibrium strategy of the other player. We present a full characterization of safe strategies, and develop efficient algorithms for exploiting suboptimal opponents while guaranteeing safety. We also provide analogous results for sequential perfect and imperfect-information games, and present safe exploitation algorithms and full characterizations of safe strategies for those settings as well. We present experimental results in Kuhn poker, a canonical test problem for game-theoretic algorithms. Our experiments show that 1) aggressive safe exploitation strategies significantly outperform adjusting the exploitation within equilibrium strategies and 2) all the safe exploitation strategies significantly outperform a (non-safe) best response strategy against strong dynamic opponents.

Four desiderata for automated market makers have appeared in the literature: (1) bounded loss, (2) the ability to make a profit, (3) a vanishing bid/ask spread, and (4) unlimited market depth. Intriguingly, market makers that satisfy any three of these desiderata have appeared in the literature. However, it was an open question as to whether a market maker can simultaneously satisfy all four because the qualities are oppositional. In this paper, we design market makers that satisfy all four. We achieve this by introducing a new, practical framework. It extends constant-utility cost functions with two separate functions that are added to the prices quoted to the trader. The liquidity function uses its proceeds to increase the amount of liquidity provided by the market maker. The profit function represents a "lockbox" of savings that is separate from the rest of the market maker's decision-making process.

Abstraction followed by equilibrium finding has emerged as the leading approach to solving games. Lossless abstraction typically yields games that are still too large to solve, so lossy abstraction is needed. Unfortunately, prior lossy game abstraction algorithms have no guarantees on solution quality. We developed a framework that enables the design of lossy game abstraction algorithms with guarantees on solution quality. It simultaneously handles state and action abstraction. We define a measure of reward approximation error and transition probability error achieved by state and action abstraction in stochastic games such that the regret of the equilibrium found in the abstract game when implemented in the original, unabstracted game is upper-bounded by a function of those measures. We then develop the first lossy game abstraction algorithms with bounds on solution quality. Both of them work level-by-level up from the end of the game. One of the algorithms is greedy and the other is an integer linear program. We also prove that the abstraction problem is NP-complete (even with just action abstraction, 2 agents, and a 1-step game), but point out that this does not mean that the game abstraction problems that occur in practice cannot be solved quickly.

The VCG mechanism is the gold standard for combinatorial auctions (CAs), and it maximizes social welfare. In contrast, the revenue-maximizing (aka optimal) CA is unknown, and designing one is NP-hard. Therefore, research on optimal CAs has progressed into special settings. Notably, Levin [1997] derived the optimal CA for complements when each agent’s private type is one-dimensional. (This does not fall inside the well-studied “singleparameter environment”.) We introduce a new research avenue for increasing revenue where we poke holes in the allocation space—based on the bids—and then use a welfare-maximizing allocation rule within the remaining allocation set. In this paper, the first step down this avenue, we introduce a new form of “reserve pricing” into CAs. We show that Levin’s optimal revenue can be 2-approximated by using “monopoly reserve prices” to curtail the allocation set, followed by welfare-maximizing allocation and Levin’s payment rule. A key lemma of potential independent interest is that the expected revenue from any truthful allocation-monotonic mechanism equals the expected virtual valuation; this generalizes Myerson’s lemma [1981] from the single-parameter environment. Our mechanism is close to the gold standard and thus easier to adopt than Levin’s. It also requires less information about the prior over the bidders’ types, and is always more efficient. Finally, we show that the optimal revenue can be 6- approximated even if the “reserve pricing” is required to be symmetric across bidders.

Using a model of agent behavior based around envy-reducing strategies, we describe an iterated combinatorial auction in which the allocation and prices converge to a solution in the core of the agents' true valuations. In each round of the iterative auction mechanism, agents act on envy quotes produced by the mechanism: hints that suggest the prices of the bundles they are interested in. We describe optimal methods of generating envy quotes for various core-selecting mechanisms. Prior work on core-selecting combinatorial auctions has required agents to have perfect information about every agent's valuations to achieve a solution in the core. In contrast, here a core solution is reached even in the private information setting.

In many auctions, agents bid more aggressively than self-interest would prescribe. This can be explained by spite, where the agent's utility not only increases in the agent's surplus but also decreases as the other bidders' surpluses increase. Spite can stem from long-term benefits from making competitors worse off and from inherent psychological effects. There have been important recent game-theoretic analyses of spiteful bidding assuming all agents are equally spiteful. We present, to our knowledge, the first auction analysis in the more realistic setting where bidders may be spiteful to different extents. We show that the equilibrium bidding function can still be written in the same form — except that the spite factor is replaced by an expressed spite factor. This leads to bidders expressing spites that are higher or lower than their true spite depending on others' spite. Perhaps surprisingly, in the two-bidder case, the mapping from true spite to expressed spite is the same across all common auction mechanisms. Furthermore, even with two bidders, important properties of symmetric-spite settings cease to hold: the allocation can be inefficient and the revenue ranking may reverse between first- and second-price auctions. We also show that in sealed-bid auctions under asymmetric valuation distributions, there can be a "bargaining problem" in selecting bids. Finally, we study the generalization where agents can have different extents of spite toward different other bidders.

The use of simple auction mechanisms like the GSP in online advertising can lead to significant loss of efficiency and revenue when advertisers have rich preferences — even simple forms of expressiveness like budget constraints can lead to suboptimal outcomes. While the optimal allocation of inventory can provide greater efficiency and revenue, natural formulations of the underlying optimization problems grow exponentially in the number of features of interest, presenting a key practical challenge. To address this problem, we propose a means for automatically partitioning inventory into abstract channels so that the least relevant features are ignored. Our approach, based on LP/MIP column and constraint generation, dramatically reduces the size of the problem, thus rendering optimization computationally feasible at practical scales. Our algorithms allow for principled tradeoffs between tractability and solution quality. Numerical experiments demonstrate the computational practicality of our approach as well as the quality of the resulting abstractions.

We designed and built the Gates Hillman Prediction Market (GHPM) to predict the opening day of the Gates and Hillman Centers, the new computer science buildings at Carnegie Mellon University. The market ran for almost a year and attracted 169 active traders who placed almost 40,000 bets with an automated market maker. Ranging over 365 possible opening days, the market's event partition size is the largest ever elicited in any prediction market by an order of magnitude. A market of this size required new advances, including a novel span-based elicitation interface. The results of the GHPM are important for two reasons. First, we uncovered two flaws of current automated market makers: spikiness and liquidity-insensitivity, and we develop the mathematical underpinnings of these flaws. Second, the market provides a valuable corpus of identity-linked trades. We use this data set to explore whether the market reacted to or anticipated official communications, how self-reported trader confidence had little relation to actual performance, and how trade frequencies suggest a power law distribution. Most significantly, the data enabled us to evaluate two competing hypotheses about how markets aggregate information, the Marginal Trader Hypothesis and the Hayek Hypothesis; the data strongly support the former.

Current automated market makers over binary events suffer from two problems that make them impractical. First, they are unable to adapt to liquidity, so trades cause prices to move the same amount in both thick and thin markets. Second, under normal circumstances, the market maker runs at a deficit. In this paper, we construct a market maker that is both sensitive to liquidity and can run at a profit. Our market maker has bounded loss for any initial level of liquidity and, as the initial level of liquidity approaches zero, worst-case loss approaches zero. For any level of initial liquidity we can establish a boundary in market state space such that, if the market terminates within that boundary, the market maker books a profit regardless of the realized outcome. Furthermore, we provide guidance as to how our market maker can be implemented over very large event spaces through a novel cost-function-based sampling method

We provide a series of algorithms demonstrating that solutions according to the fundamental game-theoretic solution concept of closed under rational behavior (CURB) sets in two-player, normal-form games can be computed in polynomial time (we also discuss extensions to n-player games). First, we describe an algorithm that identifies all of a player’s best responses conditioned on the belief that the other player will play from within a given subset of its strategy space. This algorithm serves as a subroutine in a series of polynomial-time algorithms for finding all minimal CURB sets, one minimal CURB set, and the smallest minimal CURB set in a game. We then show that the complexity of finding a Nash equilibrium can be exponential only in the size of a game’s smallest CURB set. Related to this, we show that the smallest CURB set can be an arbitrarily small portion of the game, but it can also be arbitrarily larger than the supports of its only enclosed Nash equilibrium. We test our algorithms empirically and find that most commonly studied academic games tend to have either very large or very small minimal CURB sets.

No abstract available.

We investigate whether it is possible to preserve privacy in sealed-bid auctions to a maximal extent. In particular, this paper focuses on unconditional full privacy , i.e., privacy that relies neither on trusted third parties (like auctioneers), nor on computational intractability assumptions (like the hardness of factoring). These constraints imply a scenario in which bidders exchange messages according to some predefined protocol in order to jointly determine the auction outcome without revealing any additional information. It turns out that the first-price sealed-bid auction can be emulated by an unconditionally fully private protocol. However, the protocol's round complexity is exponential in the bid size, and there is no more efficient protocol. On the other hand, we prove the impossibility of privately emulating the second-price sealed-bid auction for more than two bidders. This impossibility holds even when relaxing various privacy constraints such as allowing the revelation of all but one losing bid (while maintaining anonymity) or allowing the revelation of the second highest bidder's identity.

Sourcing professionals buy several trillion dollars worth of goods and services yearly. We introduced a new paradigm called expressive commerce and applied it to sourcing. It combines the advantages of highly expressive human negotiation with the advantages of electronic reverse auctions. The idea is that supply and demand are expressed in drastically greater detail than in traditional electronic auctions, and are algorithmically cleared. This creates a Pareto efficiency improvement in the allocation (a win-win between the buyer and the sellers) but the market clearing problem is a highly complex combinatorial optimization problem. We developed the world's fastest tree search algorithms for solving it. We have hosted $35 billion of sourcing using the technology, and created $4.4 billion of hard-dollar savings plus numerous harder-to-quantify benefits. The suppliers also benefited by being able to express production efficiencies and creativity, and through exposure problem removal. Supply networks were redesigned, with quantitative understanding of the tradeoffs, and implemented in weeks instead of months.

In barter-exchange markets, agents seek to swap their items with one another, in order to improve their own utilities. These swaps consist of cycles of agents, with each agent receiving the item of the next agent in the cycle. We focus mainly on the upcoming national kidney-exchange market, where patients with kidney disease can obtain compatible donors by swapping their own willing but incompatible donors. With over 70,000 patients already waiting for a cadaver kidney in the US, this market is seen as the only ethical way to significantly reduce the 4,000 deaths per year attributed to kidney diseas.The clearing problem involves finding a social welfare maximizing exchange when the maximum length of a cycle is fixed. Long cycles are forbidden, since, for incentive reasons, all transplants in a cycle must be performed simultaneously. Also, in barter-exchanges generally, more agents are affected if one drops out of a longer cycle. We prove that the clearing problem with this cycle-length constraint is NP-hard. Solving it exactly is one of the main challenges in establishing a national kidney exchange.We present the first algorithm capable of clearing these markets on a nationwide scale. The key is incremental problem formulation. We adapt two paradigms for the task: constraint generation and column generation. For each, we develop techniques that dramatically improve both runtime and memory usage. We conclude that column generation scales drastically better than constraint generation. Our algorithm also supports several generalizations, as demanded by real-world kidney exchanges.Our algorithm replaced CPLEX as the clearing algorithm of the Alliance for Paired Donation, one of the leading kidney exchanges. The match runs are conducted every two weeks and transplants based on our optimizations have already been conducted.

In multiagent settings where the agents have different preferences, preference aggregation is a central issue. Voting is a general method for preference aggregation, but seminal results have shown that all general voting protocols are manipulable. One could try to avoid manipulation by using protocols where determining a beneficial manipulation is hard. Especially among computational agents, it is reasonable to measure this hardness by computational complexity. Some earlier work has been done in this area, but it was assumed that the number of voters and candidates is unbounded. Such hardness results lose relevance when the number of candidates is small, because manipulation algorithms that are exponential only in the number of candidates (and only slightly so) might be available. We give such an algorithm for an individual agent to manipulate the Single Transferable Vote (STV) protocol, which has been shown hard to manipulate in the above sense. This motivates the core of this article, which derives hardness results for realistic elections where the number of candidates is a small constant (but the number of voters can be large). The main manipulation question we study is that of coalitional manipulation by weighted voters. (We show that for simpler manipulation problems, manipulation cannot be hard with few candidates.) We study both constructive manipulation (making a given candidate win) and destructive manipulation (making a given candidate not win). We characterize the exact number of candidates for which manipulation becomes hard for the plurality , Borda , STV , Copeland , maximin , veto , plurality with runoff , regular cup , and randomized cup protocols. We also show that hardness of manipulation in this setting implies hardness of manipulation by an individual in unweighted settings when there is uncertainty about the others' votes (but not vice-versa). To our knowledge, these are the first results on the hardness of manipulation when there is uncertainty about the others' votes.

Finding an equilibrium of an extensive form game of imperfect information is a fundamental problem in computational game theory, but current techniques do not scale to large games. To address this, we introduce the ordered game isomorphism and the related ordered game isomorphic abstraction transformation . For a multi-player sequential game of imperfect information with observable actions and an ordered signal space, we prove that any Nash equilibrium in an abstracted smaller game, obtained by one or more applications of the transformation, can be easily converted into a Nash equilibrium in the original game. We present an algorithm, GameShrink , for abstracting the game using our isomorphism exhaustively. Its complexity is õ ( n 2 ), where n is the number of nodes in a structure we call the signal tree. It is no larger than the game tree, and on nontrivial games it is drastically smaller, so GameShrink has time and space complexity sublinear in the size of the game tree. Using GameShrink , we find an equilibrium to a poker game with 3.1 billion nodes—over four orders of magnitude more than in the largest poker game solved previously. To address even larger games, we introduce approximation methods that do not preserve equilibrium, but nevertheless yield ( ex post ) provably close-to-optimal strategies.

In multiagent systems, strategic settings are often analyzed under the assumption that the players choose their strategies simultaneously. However, this model is not always realistic. In many settings, one player is able to commit to a strategy before the other player makes a decision. Such models are synonymously referred to as leadership, commitment, or Stackelberg models, and optimal play in such models is often significantly different from optimal play in the model where strategies are selected simultaneously.The recent surge in interest in computing game-theoretic solutions has so far ignored leadership models (with the exception of the interest in mechanism design, where the designer is implicitly in a leadership position). In this paper, we study how to compute optimal strategies to commit to under both commitment to pure strategies and commitment to mixed strategies, in both normal-form and Bayesian games. We give both positive results (efficient algorithms) and negative results (NP-hardness results).

Finding an equilibrium of an extensive form game of imperfect information is a fundamental problem in computational game theory, but current techniques do not scale to large games. To address this, we introduce the ordered game isomorphism and the related ordered game isomorphic abstraction transformation. For a multi-player sequential game of imperfect information with observable actions and an ordered signal space, we prove that any Nash equilibrium in an abstracted smaller game, obtained by one or more applications of the transformation, can be easily converted into a Nash equilibrium in the original game. We present an algorithm, GameShrink, for abstracting the game using our isomorphism exhaustively. Its complexity is Õ(n2), where n is the number of nodes in a structure we call the signal tree. It is no larger than the game tree, and on nontrivial games it is drastically smaller, so GameShrink has time and space complexity sublinear in the size of the game tree. Using GameShrink, we find an equilibrium to a poker game with 3.1 billion nodes--over four orders of magnitude more than in the largest poker game solved previously. We discuss several electronic commerce applications for GameShrink. To address even larger games, we introduce approximation methods that do not preserve equilibrium, but nevertheless yield (ex post) provably close-to-optimal strategies.

In this article, we study the problem of online market clearing where there is one commodity in the market being bought and sold by multiple buyers and sellers whose bids arrive and expire at different times. The auctioneer is faced with an online clearing problem of deciding which buy and sell bids to match without knowing what bids will arrive in the future. For maximizing profit , we present a (randomized) online algorithm with a competitive ratio of ln( p max − p min ) + 1, when bids are in a range [ p min , p max ], which we show is the best possible. A simpler algorithm has a ratio twice this, and can be used even if expiration times are not known. For maximizing the number of trades, we present a simple greedy algorithm that achieves a factor of 2 competitive ratio if no money-losing trades are allowed. We also show that if the online algorithm is allowed to subsidize matches---match money-losing pairs if it has already collected enough money from previous pairs to pay for them---then it can actually be 1-competitive with respect to the optimal offline algorithm that is not allowed subsidy. That is, for maximizing the number of trades, the ability to subsidize is at least as valuable as knowing the future. We also consider objectives of maximizing buy or sell volume and social welfare. We present all of these results as corollaries of theorems on online matching in an incomplete interval graph.We also consider the issue of incentive compatibility, and develop a nearly optimal incentive-compatible algorithm for maximizing social welfare. For maximizing profit , we show that no incentive-compatible algorithm can achieve a sublinear competitive ratio, even if only one buy bid and one sell bid are alive at a time. However, we provide an algorithm that, under certain mild assumptions on the bids, performs nearly as well as the best fixed pair of buy and sell prices, a weaker but still natural performance measure. This latter result uses online learning methods, and we also show how such methods can be used to improve our “optimal” algorithms to a broader notion of optimality. Finally, we show how some of our results can be generalized to settings in which the buyers and sellers themselves have online bidding strategies, rather than just each having individual bids.

We determine the communication complexity of the common voting rules. The rules (sorted by their communication complexity from low to high) are plurality, plurality with runoff, single transferable vote (STV), Condorcet, approval, Bucklin, cup, maximin, Borda, Copeland, and ranked pairs. For each rule, we first give a deterministic communication protocol and an upper bound on the number of bits communicated in it; then, we give a lower bound on (even the nondeterministic) communication requirements of the voting rule. The bounds match for all voting rules except STV and maximin.

We study various computational aspects of solving games using dominance and iterated dominance. We first study both strict and weak dominance (not iterated), and show that checking whether a given strategy is dominated by some mixed strategy can be done in polynomial time using a single linear program solve. We then move on to iterated dominance. We show that determining whether there is some path that eliminates a given strategy is NP-complete with iterated weak dominance. This allows us to also show that determining whether there is a path that leads to a unique solution is NP-complete. Both of these results hold both with and without dominance by mixed strategies. (A weaker version of the second result (only without dominance by mixed strategies) was already known [7].) Iterated strict dominance, on the other hand, is path-independent (both with and without dominance by mixed strategies) and can therefore be done in polynomial time.We then study what happens when the dominating strategy is allowed to place positive probability on only a few pure strategies. First, we show that finding the dominating strategy with minimum support size is NP-complete (both for strict and weak dominance). Then, we show that iterated strict dominance becomes path-dependent when there is a limit on the support size of the dominating strategies, and that deciding whether a given strategy can be eliminated by iterated strict dominance under this restriction is NP-complete (even when the limit on the support size is 3).Finally, we study Bayesian games. We show that, unlike in normal form games, deciding whether a given pure strategy is dominated by another pure strategy in a Bayesian game is NP-complete (both with strict and weak dominance); however, deciding whether a strategy is dominated by some mixed strategy can still be done in polynomial time with a single linear program solve (both with strict and weak dominance). Finally, we show that iterated dominance using pure strategies can require an exponential number of iterations in a Bayesian game (both with strict and weak dominance).

A fast-growing body of research in the AI and machine learning communities addresses learning in games, where there are multiple learners with different interests. This research adds to more established research on learning in games conducted in economics. In part because of a clash of fields, there are widely varying requirements on learning algorithms in this domain. The goal of this paper is to demonstrate how communication complexity can be used as a lower bound on the required learning time or cost. Because this lower bound does not assume any requirements on the learning algorithm, it is universal, applying under any set of requirements on the learning algorithm.We characterize exactly the communication complexity of various solution concepts from game theory, namely Nash equilibrium, iterated dominant strategies (both strict and weak), and backwards induction. This gives the tighest lower bounds on learning in games that can be obtained with this method.

When donating money to a (say, charitable) cause, it is possible touse the contemplated donation as negotiating material to induce other parties interested in the charity to donate more. Such negotiation is usually done in terms of matching offers, where one party promises to pay a certain amount if others pay a certain amount. However, in their current form, matching offers allow for only limited negotiation. For one, it is not immediately clear how multiple parties can make matching offers at the same time without creating circular dependencies. Also, it is not immediately clear how to make adonation conditional on other donations to multiple charities, when the donator has different levels of appreciation for the different charities. In both these cases, the limited expressiveness of matching offers causes economic loss: it may happen that an arrangement that would have made all parties (donators as well as charities) better off cannot be expressed in terms of matching offers and will therefore notoccur.In this paper, we introduce a bidding language for expressing very general types of matching offers over multiple charities. We formulate the corresponding clearing problem (deciding how much each bidder pays, and how much each charity receives), and show that it is NP-complete to approximate to any ratio even in very restricted settings. We givea mixed-integer program formulation of the clearing problem, and show that for concave bids, the program reduces to a linear program. We then show that the clearing problem for a subclass of concave bids is at least as hard as the decision variant of linear programming. Subsequently, we show that the clearing problem is much easier when bids are quasilinear---for surplus, the problem decomposes across charities, and for payment maximization, a greedy approach isoptimal if the bids are concave (although this latter problem is weakly NP-complete when the bids are not concave). For the quasilinear setting, we study the mechanism design question. We show that anex-post efficient mechanism is impossible even with only one charity and a very restricted class of bids. We also show that there may bebenefits to linking the charities from a mechanism design stand point.

Often, an outcome must be chosen on the basis of the preferences reported by a group of agents. The key difficulty is that the agents may report their preferences insincerely to make the chosen outcome more favorable to themselves. Mechanism design is the art of designing the rules of the game so that the agents are motivated to report their preferences truthfully, and a desirable outcome is chosen. In a recently proposed approach---called automated mechanism design ---a mechanism is computed for the preference aggregation setting at hand. This has several advantages, but the downside is that the mechanism design optimization problem needs to be solved anew each time. Unlike the earlier work on automated mechanism design that studied a benevolent designer, in this paper we study automated mechanism design problems where the designer is self-interested. In this case, the center cares only about which outcome is chosen and what payments are made to it. The reason that the agents' preferences are relevant is that the center is constrained to making each agent at least as well off as the agent would have been had it not participated in the mechanism. In this setting, we show that designing optimal deterministic mechanisms is NP-complete in two important special cases: when the center is interested only in the payments made to it, and when payments are not possible and the center is interested only in the outcome chosen. We then show how allowing for randomization in the mechanism makes problems in this setting computationally easy. Finally, we show that the payment-maximizing AMD problem is closely related to an interesting variant of the optimal (revenue-maximizing) combinatorial auction design problem, where the bidders have "best-only" preferences. We show that here, too, designing an optimal deterministic auction is NP-complete, but designing an optimal randomized auction is easy.

No abstract available.

No abstract available.

In many market settings, agents do not know their preferences a priori. Instead, they may have to solve computationally complex optimization problems, query databases, or perform expensive searches in order to determine their values for di#erent outcomes. For such settings, we have introduced the deliberation equilibrium as the game-theoretic solution concept where the agents' deliberation actions are modeled as part of their strategies.

We study auctioning multiple units of the same good to potential buyers with single unit demand (i.e. every bidder wants only one unit of the good). Depending on the objective of the seller, different selling mechanisms are desirable. The Vickrey auction with a truthful reserve price is optimal when the objective is efficiency - allocating the units to the parties who values them the most. The Myerson auction is optimal when the objective is the seller's expected utility. These two objectives are generally in conflict, and cannot be maximized with one mechanism. In many real-world settings---such as privatization and competing electronic marketplaces---it is not clear that the objective should be either efficiency or seller's expected utility. Typically, one of these objectives should weigh more than the other, but both are important. We account for both objectives by designing a new deterministic dominant strategy auction mechanism that maximizes expected social welfare subject to aminimum constraint on the seller's expected utility. This way the seller can maximize social welfare subject to doing well enough for himself.

The aggregation of conflicting preferences is a central problem in multiagent systems. The key difficulty is that the agents may report their preferences insincerely. Mechanism design is the art of designing the rules of the game so that the agents are motivated to report their preferences truthfully and a (socially) desirable outcome is chosen. We propose an approach where a mechanism is automatically created for the preference aggregation setting at hand. This has several advantages, but the downside is that the mechanism design optimization problem needs to be solved anew each time. Hence the computational complexity of mechanism design becomes a key issue. In this paper we analyze the single-agent mechanism design problem, whose simplicity allows for elegant and generally applicable results.We show that designing an optimal deterministic mechanism that does not use payments is NP-complete even if there is only one agent whose type is private information---even when the designer's objective is social welfare. We show how this hardness result extends to settings with multiple agents with private information. We then show that if the mechanism is allowed to use randomization, the design problem is solvable by linear programming (even for general objectives) and hence in P. This generalizes to any fixed number of agents. We then study settings where side payments are possible and the agents' preferences are quasilinear. We show that if the designer's objective is social welfare, an optimal deterministic mechanism is easy to construct; in fact, this mechanism is also ex post optimal. We then show that designing an optimal deterministic mechanism with side payments is NP-complete for general objectives, and this hardness extends to settings with multiple agents. Finally, we show that an optimal randomized mechanism can be designed in polynomial time using linear programming even for general objective functions. This again generalizes to any fixed number of agents.

We study how to learn to play a Pareto-optimal strict Nash equilibrium when there exist multiple equilibria and agents may have different pref- erences among the equilibria. We focus on repeated coordination games of non-identical interest where agents do not know the game structure up front and receive noisy payoffs. We design efficient near-optimal al- gorithms for both the perfect monitoring and the imperfect monitoring setting(where the agents only observe their own payoffs and the joint actions).

Combinatorial auctions, where bidders can submit bids on bundles of items, are attractive when the bidders' valuations on bundles exhibit complementarity and/or substitutability. An important problem is that of eliciting the bidders' preferences so that they do not have to bid on all combinations. We present a new family of preference elicitation algorithms. The algorithms in this family do not rely on absolute bids, but rather on relative (differential) value information. This holds the promise to reduce revelation and communication significantly. We develop a differential-elicitation algorithm that finds the efficient allocation of items to the bidders, and as a side-effect, the Vickrey payments (which make truthful bidding incentive compatible). We also present two auction mechanisms that use differential elicitation.

Coalition formation is a key problem in automated negotiation among self-interested agents, and other electronic commerce applications. A coalition of agents can sometimes accomplish things that the individual agents cannot, or can do things more efficiently. However, motivating the agents to abide to a solution requires careful analysis: only some of the solutions are stable in the sense that no group of agents is motivated to break off and form a new coalition. This constraint has been studied extensively in cooperative game theory. However, the computational questions around this constraint have received less attention. When it comes to coalition formation among software agents (that represent real-world parties), these questions become increasingly explicit.In this paper we define a concise general representation for games in characteristic form that relies on superadditivity, and show that it allows for efficient checking of whether a given outcome is in the core. We then show that determining whether the core is nonempty is NP-complete both with and without transferable utility. We demonstrate that what makes the problem hard in both cases is determining the collaborative possibilities (the set of outcomes possible for the grand coalition), by showing that if these are given, the problem becomes tractable in both cases. However, we then demonstrate that for a hybrid version of the problem, where utility transfer is possible only within the grand coalition, the problem remains NP-complete even when the collaborative possibilities are given.

Often, an outcome must be chosen on the basis of the preferences reported by a group of agents. The key difficulty is that the agents may report their preferences insincerely to make the chosen outcome more favorable to themselves. Mechanism design is the art of designing the rules of the game so that the agents are motivated to report their preferences truthfully, and a desirable outcome is chosen. We recently proposed an approach---called automated mechanism design---where a mechanism is computed for the preference aggregation setting at hand. This has several advantages, but the downside is that the mechanism design optimization problem needs to be solved anew each time. Unlike the earlier work on automated mechanism design that studied a benevolent designer, in this paper we study automated mechanism design problems where the designer is self-interested. In this case, the center cares only about which outcome is chosen and what payments are made to it. The reason that the agents' preferences are relevant is that the center is constrained to making each agent at least as well off as the agent would have been had it not participated in the mechanism. In this setting, we show that designing optimal deterministic mechanisms is NP-complete in two important special cases: when the center is interested only in the payments made to it, and when payments are not possible and the center is interested only in the outcome chosen. We then show how allowing for randomization in the mechanism makes problems in this setting computationally easy.

Combinatorial auctions, where bidders can bid on bundles of items, are known to be desirable auction mechanisms for selling items that are complementary and/or substitutable. However, there are 2k --1 bundles, and each agent may need to bid on all of them to fully express its preferences. We address this by showing how them auctioneer can recommend to the agents incrementally which bundles to bid on so that they need to only place a small fraction of all possible bids. These algorithms impose a great computational burden on the auctioneer; we show how to speed them up dramatically. We also present an optimal elicitor, which is intractable but may be the basis for future algorithms. Finally, we introduce the notion of a universal revelation reducer, demonstrate a randomized one, and prove that no deterministic one exists.The full paper is available in draft form at http://www.cs.cmu.edu/ sandholm/using_value_queries.pdf.

No abstract available.

We study selling one indivisible object to multiple potential buyers. Depending on the objective of the seller, different selling mechanisms are desirable. The Vickrey auction with a truthful reserve price is optimal when the objective is efficiency (i.e., allocating the object to the party who values it the most). The Myerson auction is optimal when the objective is the seller's expected utility. These two objectives are generally in conflict, and cannot be maximized with one mechanism. In many real-world settings---such as privatization and competing electronic marketplaces---it is not clear that the objective should be either efficiency or seller's expected utility. Typically, one of these objectives should weigh more than the other, but both are important. We account for importance of both objectives by designing a new deterministic auction mechanism that maximizes expected social welfare subject to a minimum constraint on the seller's expected utility. This way the seller can expect to do well enough for himself, while maintaining the attractive properties of the mechanism.

Preference elicitation --- the process of asking queries to determine parties' preferences --- is a key part of many problems in electronic commerce. For example, a shopping agent needs to know a user's preferences in order to correctly act on her behalf, and preference elicitation can help an auctioneer in a combinatorial auction determine how to best allocate a given set of items to a given set of bidders. Unfortunately, in the worst case, preference elicitation can require an exponential number of queries even to determine an approximately optimal allocation. In this paper we study natural special cases of preferences for which elicitation can be done in polynomial time via value queries. The cases we consider all have the property that the preferences (or approximations to them) can be described in a polynomial number of bits, but the issue here is whether they can be elicited using the natural (limited) language of value queries. We make a connection to computational learning theory where the similar problem of exact learning with membership queries has a long history. In particular, we consider preferences that can be written as read-once formulas over a set of gates motivated by a shopping application, as well as a class of preferences we call Toolbox DNF, motivated by a type of combinatorial auction. We show that in each case, preference elicitation can be done in polynomial time. We also consider the computational problem of allocating items given the parties' preferences, and show that in certain cases it can be done in polynomial time and in other cases it is NP-complete. Given two bidders with Toolbox-DNF preferences, we show that allocation can be solved via network flow. If parties have read-once formula preferences, then allocation is NP-hard even with ju...

In multiagent settings where the agents have different preferences, preference aggregation is a central issue. Voting is a general method for preference aggregation, but seminal results have shown that all general voting protocols are manipulable. One could try to avoid manipulation by using voting protocols where determining a beneficial manipulation is hard computationally. The complexity of manipulating realistic elections where the number of candidates is a small constant was recently studied [4], but the emphasis was on the question of whether or not a protocol becomes hard to manipulate for some constant number of candidates. That work, in many cases, left open the question: How many candidates are needed to make elections hard to manipulate? This is a crucial question when comparing the relative manipulability of different voting protocols. In this paper we answer that question for the voting protocols of the earlier study: plurality, Borda, STV, Copeland, maximin, regular cup, and randomized cup. We also answer that question for two voting protocols for which no results on the complexity of manipulation have been derived before: veto and plurality with runoff. It turns out that the voting protocols under study become hard to manipulate at 3 candidates, 4 candidates, 7 candidates, or never.

Multiagent learning is a key problem in AI. In the presence of multi- ple Nash equilibria, even agents with non-conflicting interests may not be able to learn an optimal coordination policy. The problem is exac- cerbated if the agents do not know the game and independently receive noisy payoffs. So, multiagent reinforfcement learning involves two inter- related problems: identifying the game and learning to play. In this paper, we present optimal adaptive learning, the first algorithm that converges to an optimal Nash equilibrium with probability 1 in any team Markov game. We provide a convergence proof, and show that the algorithm’s parameters are easy to set to meet the convergence conditions.

The paper investigates the impact of trust on market efficiency and bilateral contracts. We prove that a market in which agents are trusted to the degree they deserve to be trusted is as efficient as a market with complete trustworthiness. In other words, complete trustworthiness is not a necessary condition for market efficiency. We prove that distrust could significantly reduce market efficiency, and we show how to solve the problem by using appropriately designed multiagent contracts. The problem of trust is studied in the context of a bilateral negotiation game between a buyer and a seller. It is shown that if the seller’s trust equals the buyer’s trustworthiness, then the social welfare, the amount of trade, and the agents’ utility functions are maximized. The paper also studies the efficiency of advance payment contracts as a tool for improving trustworthiness. It is proved that advance payment contracts maximize the social welfare and the amount of trade. Finally, the paper studies the problem of how to make agents truthfully reveal their level of trustworthiness. An incentive–compatible contract is defined, in which agents do not benefit from lying about their trustworthiness. The analysis and the solutions proposed in this paper could help agent designers avoid many market failures and produce efficient interaction mechanisms.

This paper presents eMediator, an electronic commerce server prototype that demonstrates ways in which algorithmic support and game–theoretic incentive engineering can jointly improve the efficiency of e–commerce. eAuctionHouse, the configurable auction server, includes a variety of generalized combinatorial auctions and exchanges, pricing schemes, bidding languages, mobile agents, and user support for choosing an auction type. We introduce two new logical bidding languages for combinatorial markets: the XOR bidding language and the OR–of–XORs bidding language. Unlike the traditional OR bidding language, these are fully expressive. They therefore enable the use of the Clarke–Groves pricing mechanism for motivating the bidders to bid truthfully. eAuctionHouse also supports supply/demand curve bidding. eCommitter, the leveled commitment contract optimizer, determines the optimal contract price and decommitting penalties for a variety of leveled commitment contracting mechanisms, taking into account that rational agents will decommit strategically in Nash equilibrium. It also determines the optimal decommitting strategies for any given leveled commitment contract. eExchangeHouse, the safe exchange planner, enables unenforced anonymous exchanges by dividing the exchange into chunks and sequencing those chunks to be delivered safely in alternation between the buyer and the seller.

Combinatorial auctions (CAs) where bidders can bid on bundles of items can be very desirable market mechanisms when the items sold exhibit complementarity and/or substitutability, so the bidder's valuations for bundles are not additive. However, in a basic CA, the bidders may need to bid on exponentially many bundles, leading to difficulties in determining those valuations, undesirable information revelation, and unnecessary communication. In this paper we present a design of an auctioneer agent that uses topological structure inherent in the problem to reduce the amount of information that it needs from the bidders. An analysis tool is presented as well as data structures for storing and optimally assimilating the information received from the bidders. Using this information, the agent then narrows down the set of desirable (welfare-maximizing or Pareto-efficient) allocations, and decides which questions to ask next. Several algorithms are presented that ask the bidders for value, order, and rank information. A method is presented for making the elicitor incentive compatible.

The capability to reallocate items-e.g. tasks, securities, bandwidth slices, MW hours of electricity, and collectibles-is a key feature in automated negotiation. Especially when agents have preferences over combinations of items, this is highly nontrivial. Marginal cost based reallocation leads to an anytime algorithm where every agent's utility increases monotonically over time. Different contract types head toward different locally optimal task allocations, and contracts from a recently introduced comprehensive contract type, OCSM-contracts, head toward the global optimum. Reaching it can take an impractically long time, so it is important to trade off solution quality against negotiation time. To construct negotiation protocols that lead to the best achievable allocations in a bounded amount of time, we compared sequences of four contract types. Original, cluster, swap, and multiagent contracts. The experiments show that it is profitable to use multiple contract types in the sequence: significantly better solutions are reached, and faster, than if only one contract type is used. However, the best sequences only include original and cluster contracts. Swap and multiagent contracts lead to bad local optima quickly. Interestingly, the number of contracts using any given contract type does not always decrease over time: contracts play the role of enabling further contracts.

Coalition formation is a key topic in multiagent systems. One would prefer a coalition structure that maximizes the sum of the values of the coalitions, but often the number of coalition structures is too large to allow for exhaustive search for the optimal one. We present experimental results for three anytime algorithms that search the space of coalition structures. We show that, in the average case, all three algorithms do much better than the recently established theoretical worst case results in Sandholm et al. (1999a). We also show that no one algorithm is dominant. Each algorithm's performance is influenced by the particular instance distribution, with each algorithm outperforming the others for different instances. We present a possible explanation for the behaviour of the algorithms and support our hypothesis with data collected from a controlled experimental run.

The paper investigates the impact of trust on the form and the terms of contracts.A formal model of a bilateral game between a buyer and a seller is presented.The paper focuses on the case where the level of trustworthiness of the buyer is uncertain.It is shown that if the seller's trust equals the buyer trustworthiness, then the social welfare, the amount of trade and the agents' utility functions are maximized.It is demonstrated that underestimating the buyer's trustworthiness tends to harm both agents.The role of advance payment contracts for improving trustworthiness is discussed.It is shown that in the case where the buyer is distrusted, advance payment contracts are efficient in terms of amount of trade and social welfare.It is proved that in this case both agents prefer an advance payment contract to a standard uncertain payment contract.Advance payment F-contracts are proposed.It is proved that these contracts are incentive compatible and that they eliminate inefficiencies caused by asymmetric information about trust.

This paper studies coalition formation among self-interested agents that cannot make sidepayments. We show that α-core stability reduces to analysing whether some utility profile is maximal for all agents. We also show that strategy profiles that lead to the α-core are a subset of Strong Nash equilibria. This fact carries our α-core-based stability results directly over to two other strategic solution concepts: Nash equilibrium and Coalition-Proof Nash equilibrium. The main focus of the paper is to analyse the dynamic process of coalition formation by explicitly modelling the costs of communication and deliberation. We describe an algorithm for sequential action choice where each agent greedily stepwise maximizes its payoff given its beliefs. Conditions are derived under which this process leads to convergence of the agents' beliefs and to a stable coalition structure. We derive these results for the case where the length of the process is exogenously restricted as well as for the case where agents can choose it. Finally, we show that the outcome of any communication/deliberation process that leads to a stable coalition structure is Pareto-optimal for the original game which does not incorporate communication or deliberation. Conversely, any Pareto-optimal outcome can be supported by a communication/deliberation process that leads to a stable coalition structure.