Online marketplaces will be transformed by autonomous AI agents acting on behalf of consumers. Rather than humans browsing and clicking, AI agents can parse webpages or interact through APIs to evaluate products, and transact. This raises a fundamental question: what do AI agents buy—and why? We develop ACES, a sandbox environment that pairs a platform-agnostic agent with a fully programmable mock marketplace to study this. We first explore aggregate choices, revealing that modal choices can differ across models, with AI agents sometimes concentrating on a few products, raising competition questions. We then analyze the current drivers of choices through randomized experiments on product positions and listing attributes. Models show sizeable and heterogeneous position effects: all favor the top row, yet different models prefer different columns, undermining the assumption of a universal ''top'' rank. They penalize sponsored tags, reward endorsements, and sensitivities to price, ratings, and reviews are directionally as expected, but vary sharply across models. Our findings reveal how AI agents behave in e-commerce, and surface concrete monitoring, seller strategy, platform design, and regulatory questions.

Electric power systems are undergoing a major transformation as they integrate intermittent renewable energy sources, and batteries to smooth out variations in renewable energy production. As privately-owned batteries grow from their role as marginal "price-takers" to significant players in the market, a natural question arises: How do batteries operate in electricity markets, and how does the strategic behavior of decentralized batteries distort decisions compared to centralized batteries? We propose an analytically tractable model that captures salient features of the highly complex electricity market. We derive in closed form the resulting battery behavior and generation cost in three operating regimes: (i) no battery, (ii) centralized battery, and (ii) decentralized profit-maximizing battery. We establish that a decentralized battery distorts its discharge decisions in three ways. First, there is quantity withholding, i.e., discharging less than centrally optimal. Second, there is a shift in participation from day-ahead to real-time, i.e., postponing some of its discharge from day-ahead to real-time. Third, there is reduction in real-time responsiveness, or discharging less in response to smoothing real-time demand than centrally optimal. We also quantify the impact of the battery market power on total system cost via the Price of Anarchy metric, and prove that it is always between 9/8 and 4/3. That is, incentive misalignment always exists, but it is bounded even in the worst case. We calibrate our model to real data from Los Angeles and Houston. Lastly, we show that competition is very effective at reducing distortions, but many market power mitigation mechanisms backfire, and lead to higher total cost. The work provides stakeholders with a framework to understand and detect market power from batteries. It also shows that the potential loss from battery market power is relatively small compared to the cost reduction achievable from having enough battery capacity in the system. Therefore, independent system operators in rapidly changing markets might want to prioritize market entry of batteries and only shift to market power mitigation once the market is more mature.

Decision-making often requires an individual to navigate a multitude of options with incomplete knowledge of their own preferences. Information provisioning tools such as public rankings and personalized recommendations have become central to helping individuals make choices, yet their value proposition under different marketplace environments remains unexplored. This paper studies a stylized model to explore the impact of these tools in two marketplace settings: uncapacitated supply, where items can be selected by any number of agents, and capacitated supply, where each item is constrained to be matched to a single agent. We model the agents utility as a weighted combination of a common term which depends only on the item, reflecting the item's population-level quality, and an idiosyncratic term, which depends on the agent-item pair capturing individual-specific preferences. Public rankings reveal the common term, while personalized recommendations reveal both terms.

Recommendation systems are widespread, and through customized recommendations, promise to match users with options they will like. To that end, data on engagement is collected and used. Most recommendation systems are ranking-based, where they rank and recommend items based on their predicted engagement. However, the engagement signals are often only a crude proxy for user utility, as data on the latter is rarely collected or available. This paper explores the following question: By optimizing for measurable proxies, are recommendation systems at risk of significantly under-delivering on user utility? If that is indeed the case, how can one improve utility which is seldom measured? To study these questions, we introduce a model of repeated user consumption in which, at each interaction, users select between an outside option and the best option from a recommendation set. Our model accounts for user heterogeneity, with the majority preferring “popular” content, and a minority favoring “niche” content. The system initially lacks knowledge of individual user preferences but can learn these preferences through observations of users’ choices over time. Our theoretical and numerical analysis demonstrate that optimizing for engagement signals can lead to significant utility losses. Instead, we propose a utility-aware policy that initially recommends a mix of popular and niche content. We show that such a policy substantially improves utility despite not measuring it. As the platform becomes more forward-looking, our utility-aware policy achieves the best of both worlds: near-optimal user utility and near-optimal engagement simultaneously. Our study elucidates an important feature of recommendation systems; given the ability to suggest multiple items, one can perform significant exploration without incurring significant reductions in short term engagement. By recommending high-risk, high-reward items alongside popular items, systems can enhance discovery of high utility items without significantly affecting engagement.

A seller wants to sell an indivisible item to n buyers. The buyer valuations are drawn i.i.d. from a distribution, but the seller does not know this distribution; the seller only knows the support [a, b]. To be robust against the lack of knowledge of the environment and buyers' behavior, the seller optimizes over dominant strategy incentive compatible (DSIC) mechanisms, and measures the worst-case performance relative to an oracle with complete knowledge of buyers' valuations. Our analysis encompasses both the regret and the approximation ratio objectives.

Platforms facilitating many-to-many matches in two-sided markets have become ubiquitous across industries ranging from professional services to dating. Differently from standard (one-sided) markets where consumers choose goods or services, in two-sided markets, both sides have preferences. Since these preferences can often be hard to describe, centralized matching is difficult to implement. The alternative option is for the platform to operate in a decentralized fashion, leaving the agents from both sides "free to find each other". While easier to implement, the downside of decentralized systems is that inefficiencies driven by congestion are likely to arise. In the present paper, we are primarily interested in understanding the power of "detail-free" levers that decentralized platforms can leverage to improve market outcomes. In particular, we focus on the lever of information design through competition signaling, where the platform discloses how much competition currently exists for a given supply unit. Signaling that there is competition for a supply unit may reduce the value of that unit but may also redirect the demand's attention to alternative supply units, potentially increasing the value for the platform. To quantify the trade-off at play and tackle the question above, we focus on a specific labor platform and the submarket of cleaning services to answer this question empirically. We partnered with the largest service labor marketplace in Latin America, which operates as follows. Service providers (agents) join the platform to purchase nonexclusive leads for jobs posted by supply-side customers. When they purchase a lead, they are not guaranteed to get the job, but simply purchase the contact information of the customer in order to apply for the job. A key characteristic of this market is the possible congestion on the lead side. In the context of such a platform, to understand the impact of any lever on market outcomes, it is fundamental to first understand how agents make their lead purchasing decisions and, in particular, how they take competition into account when making such decisions. We propose a structural model in which agents use a prediction function to forecast how much competition they may face. We show that if agents are strategic, a natural concept of equilibrium arises. By leveraging the platforms' data and an quasi-experiment, we estimate the structural parameters in the model. We find that agents react strongly to observed competition and predictions of future competition. We then conduct counterfactual analysis to study the impact of signaling competition. Our findings show that it is a powerful lever to improve market outcomes in this market. Signaling competition improves (decreases) congestion, and it also improves (increases) the probability that a lead will receive at least one applicant. Furthermore, displaying competition leads to an increase in overall leads purchased.

In this work, we study data-driven decision-making and depart from the classical identically and independently distributed (i.i.d.) assumption. We present a new framework in which historical samples are generated from unknown and different distributions, which we dub \textit{heterogeneous environments}. These distributions are assumed to lie in a heterogeneity ball with known radius and centered around the (also) unknown future (out-of-sample) distribution on which the performance of a decision will be evaluated. We quantify the asymptotic worst-case regret that is achievable by central data-driven policies such as Sample Average Approximation, but also by rate-optimal ones, as a function of the radius of the heterogeneity ball. Our work shows that the type of achievable performance varies considerably across different combinations of problem classes and notions of heterogeneity. We demonstrate the versatility of our framework by comparing achievable guarantees for the heterogeneous version of widely studied data-driven problems such as pricing, ski-rental, and newsvendor. En route, we establish a new connection between data-driven decision-making and distributionally robust optimization.

Classical Bayesian mechanism design relies on the common prior assumption, but the common prior is often not available in practice. We study the design of prior-independent mechanisms that relax this assumption: the seller is selling an indivisible item to n buyers such that the buyers' valuations are drawn from a joint distribution that is unknown to both the buyers and the seller; buyers do not need to form beliefs about competitors, and the seller assumes the distribution is adversarially chosen from a specified class. We measure performance through the worst-caseregret, or the difference between the expected revenue achievable with perfect knowledge of buyers' valuations and the actual mechanism revenue.

We study the multi-secretary problem with capacity to hire up to B out of T candidates, and values drawn i.i.d. from a distribution F on [0,1]. We investigate achievable regret performance, where the latter is defined as the difference between the performance of an oracle with perfect information of future types (values) and an online policy. While the case of distributions over a few discrete types is well understood, very little is known when there are many types, with the exception of the special case of a uniform distribution of types. In this work we consider a larger class of distributions which includes the few discrete types as a special case. We first establish the insufficiency of the common certainty equivalent heuristic for distributions with many types and "gaps" (intervals) of absent types; even for simple deviations from the uniform distribution, it leads to regret Θ(√T), as large as that of a non-adaptive algorithm. We introduce a new algorithmic principle which we call "conservativeness with respect to gaps" (CwG), and use it to design an algorithm that applies to any distribution. We establish that the proposed algorithm yields optimal regret scaling of ~Θ (T1/2 - 1/(2(β + 1))) for a broad class of distributions with gaps, where β quantifies the mass accumulation of types around gaps. We recover constant regret scaling for the special case of a bounded number of types (β=0 in this case). In most practical network revenue management problems, the number of types is large and the current certainty equivalent heuristics scale poorly with the number of types. The new algorithmic principle called Conservatism w.r.t Gaps (CwG) that we developed, can pave the way for progress on handling many types for the broader class of network revenue management problems like order fulfillment and online matching.

We study the problem of online contextual optimization where, at each period, instead of observing the loss, we observe, after-the-fact, the optimal action an oracle with full knowledge of the objective function would have taken. At each period, the decision-maker has access to a new set of feasible actions to select from and to a new contextual function that affects that period’s loss function. We aim to minimize regret, which is defined as the difference between our losses and the ones incurred by an all-knowing oracle. We obtain the first regret bound for this problem that is logarithmic in the time horizon. Our results are derived through the development and analysis of a novel algorithmic structure that leverages the underlying geometry of the problem.

We study the following fundamental data-driven pricing problem. How can/should a decision-maker price its product based on observations at a single historical price? The decision-maker optimizes over (potentially randomized) pricing policies to maximize the worst-case ratio of the revenue it can garner compared to an oracle with full knowledge of the distribution of values, when the latter is only assumed to belong to broad non-parametric set. In particular, our framework applies to the widely used regular and monotone non-decreasing hazard rate (mhr) classes of distributions. For settings where the seller knows the exact probability of sale associated with one historical price or only a confidence interval for it, we fully characterize optimal performance and near-optimal pricing algorithms that adjust to the information at hand. As examples, against mhr distributions, we show that it is possible to guarantee $85%$ of oracle performance if one knows that half of the customers have bought at the historical price, and if only $1%$ of the customers bought, it still possible to guarantee $51%$ of oracle performance. The framework we develop leads to new insights on the value of information for pricing, as well as the value of randomization. In addition, it is general and allows to characterize optimal deterministic mechanisms and incorporate uncertainty in the probability of sale.

In the present paper, we study the following fundamental problem: how should a decision-maker price based on a finite and limited number of samples from the distribution of values of customers. The decision-maker's objective is to select a pricing policy with maximum competitive ratio when the value distribution is only known to belong to some general non-parametric class. We study achievable performance for two central classes, regular and monotone hazard rate (mhr) distributions, through a general framework. To date, only results are available for a single sample and two samples. We improve existing results but also obtain the first results on achievable performance as the number of samples increases. At a higher level, this work also provides insights on the value of samples for pricing purposes. For example, against mhr distributions (resp. regular), two samples suffice to ensure 71% (resp. 61%) of optimal oracle performance, and ten samples guarantee $80%$ (resp. $65%$) of such performance. Our analysis relies on the introduction of a new (simple) class of policies and the derivation of tractable lower bounds on their performance through factor revealing dynamic programs.

One of the central challenges in online advertising is attribution, namely, assessing the contribution of individual advertiser actions including emails, display ads and search ads to eventual conversion. Several heuristics are used for attribution in practice; however, there is no formal justification for them and many of these fail even in simple canonical settings. The main contribution in this work is to develop an axiomatic framework for attribution in online advertising. In particular, we consider a Markovian model for the user journey through the conversion funnel, in which ad actions may have disparate impacts at different stages. We propose a novel attribution metric, that we refer to as counterfactual adjusted Shapley value, which inherits the desirable properties of the traditional Shapley value. Furthermore, we establish that this metric coincides with an adjusted “unique-uniform” attribution scheme. This scheme is efficiently computable and implementable and can be interpreted as a correction to the commonly used uniform attribution scheme.

Auctions are widely used in practice. While also extensively studied in the literature, most of the developments rely on significant assumptions about common knowledge on the seller and buyers' sides. In this work, we study the design of optimal prior-independent selling mechanisms. In particular, the seller faces buyers whose values are drawn from an unknown distribution, and only knows that the distribution belongs to a particular class. We analyze a competitive ratio objective, in which the seller attempts to optimize the worst-case fraction of revenues garnered compared to those of an oracle with knowledge of the distribution. Our results are along two dimensions. We first characterize the structure of optimal mechanisms. Leveraging such structure, we then establish tight lower and upper bounds on performance, leading to a crisp characterization of optimal performance for a spectrum of families of distributions. In particular, our results imply that a second price auction is an optimal mechanism when the seller only knows that the distribution of buyers has a monotone increasing hazard rate, and guarantees at least 71.53% of the optimal revenue against any distribution within this class. Furthermore, a second price auction is near-optimal when the class of admissible distributions is that of those with increasing virtual values (aka regular). Under this class, it guarantees a fraction of 50% of optimal revenues and no mechanism can guarantee more than 55.6%.

We study the dynamic mechanism design problem of a seller who repeatedly auctions independent items over a discrete time horizon to buyers who face a cumulative budget constraint. A driving motivation behind our model is the emergence of real-time bidding markets for online display advertising in which such budgets are prevalent. We assume the seller has a strong form of limited commitment: she commits to the rules of the current auction but cannot commit to those of future auctions. We show that the celebrated Myersonian approach that leverages the envelope theorem fails in this setting, and therefore, characterizing the dynamic optimal mechanism seems intractable. Despite these challenges, we derive and characterize a near-optimal dynamic mechanism. To do so, we show that the Myersonian approach is recovered in a corresponding fluid continuous time model in which the time interval between consecutive items becomes negligible. Then we leverage this approach to characterize the optimal dynamic direct-revelation mechanism, highlighting novel incentives at play in settings with buyers’ budget constraints and seller’s limited commitment. We show through a combination of theoretical and numerical results that the optimal mechanism arising from the fluid continuous time model approximately satisfies incentive compatibility for the buyers and is approximately sequentially rational for the seller in the original discrete time model. Supplemental material is available at https://doi.org/10.1287/opre.2018.1830 .

Ad Exchanges are emerging Internet markets where advertisers may purchase display ad placements, in real-time and based on specific viewer information, directly from publishers via a simple auction mechanism. Advertisers join these markets with a prespecified budget and participate in multiple second-price auctions over the length of a campaign. This paper studies the competitive landscape that arises in Ad Exchanges and the implications for publishers' decisions.