Behavioral health conditions, which include mental health and substance use disorders, are the leading disease burden in the United States. Peer-run behavioral health organizations (PROs) critically assist individuals facing these conditions by combining mental health services with assistance for needs such as income, employment, and housing. However, limited funds and staffing make it difficult for PROs to address all service user needs. To assist peer providers at PROs with their day-to-day tasks, we introduce PeerCoPilot, a large language model (LLM)-powered assistant that helps peer providers create wellness plans, construct step-by-step goals, and locate organizational resources to support these goals. PeerCoPilot ensures information reliability through a retrieval-augmented generation pipeline backed by a large database of over 1,300 vetted resources. We conducted human evaluations with 15 peer providers and 6 service users and found that over 90% of users supported using PeerCoPilot. Moreover, we demonstrate that PeerCoPilot provides more reliable and specific information than a baseline LLM. PeerCoPilot is now used by a group of 5-10 peer providers at CSPNJ, a large behavioral health organization serving over 10,000 service users, and we are actively expanding PeerCoPilot's use.

Food rescue organizations simultaneously tackle food insecurity and waste by working with volunteers to redistribute food from donors who have excess to recipients who need it. Volunteer feedback allows food rescue organizations to identify issues early and ensure volunteer satisfaction. However, food rescue organizations monitor feedback manually, which can be cumbersome and labor-intensive, making it difficult to prioritize which issues are most important. In this work, we investigate how large language models (LLMs) assist food rescue organizers in understanding and taking action based on volunteer experiences. We work with 412 Food Rescue, a large food rescue organization based in Pittsburgh, Pennsylvania, to design RescueLens, an LLM-powered tool that automatically categorizes volunteer feedback, suggests donors and recipients to follow up with, and updates volunteer directions based on feedback. We evaluate the performance of RescueLens on an annotated dataset, and show that it can recover 96% of volunteer issues at 71% precision. Moreover, by ranking donors and recipients according to their rates of volunteer issues, RescueLens allows organizers to focus on 0.5% of donors responsible for more than 30% of volunteer issues. RescueLens is now deployed at 412 Food Rescue and through semi-structured interviews with organizers, we find that RescueLens streamlines the feedback process so organizers better allocate their time.

Safe Multi-Agent Reinforcement Learning (MARL) typically relies on manually specified numeric cost functions to ensure that policy behaviours respect safety constraints. As systems scale and human-defined constraints become more diverse, context-dependent, and frequently updated, hand-crafting such cost functions becomes prohibitively complex, tedious, and error-prone. Natural language offers an intuitive and flexible alternative for defining constraints, enabling broader accessibility and easier adaptation to new scenarios and evolving rules. However, current MARL frameworks lack effective mechanisms to incorporate free-form textual constraints in a robust and principled way. To bridge this gap, we introduce Safe Multi-Agent Reinforcement Learning with natural Language constraints (SMALL), a framework that leverages fine-tuned language models to parse and encode textual constraints into semantically meaningful embeddings. These embeddings characterise prohibited states or behaviours and enable automatic prediction of constraint violations. We integrate the resulting learned costs directly into MARL training, allowing agents to optimise task performance while simultaneously minimising constraint violations, without requiring manually engineered numeric cost functions. To rigorously evaluate our method, we also propose the LaMaSafe benchmark---a set of diverse multi-agent tasks designed to assess the capability of MARL algorithms to understand and adhere to realistic, human-provided natural language constraints. Experimental results across LaMaSafe environments show that SMALL achieves comparable task performance to strong MARL baselines while significantly reducing constraint violations. While SMALL does not provide formal safety guarantees, it demonstrates that natural language can be used to shape multi-agent behaviour toward safer policies.

Peer-run organizations (PROs) provide critical, recovery-based behavioral health support rooted in lived experience. As large language models (LLMs) enter this domain, their scale, conversationality, and opacity introduce new challenges for situatedness, trust, and autonomy. Partnering with Collaborative Support Programs of New Jersey (CSPNJ), a statewide PRO in the Northeastern United States, we used comicboarding, a co-design method, to conduct workshops with 16 peer specialists and 10 service users exploring perceptions of integrating an LLM-based recommendation system into peer support. Findings show that depending on how LLMs are introduced, constrained, and co-used, they can reconfigure in-room dynamics by sustaining, undermining, or amplifying the relational authority that grounds peer support. We identify opportunities, risks, and mitigation strategies across three tensions: bridging scale and locality, protecting trust and relational dynamics, and preserving peer autonomy amid efficiency gains. We contribute design implications that center lived-experience-in-the-loop, reframe trust as co-constructed, and position LLMs not as clinical tools but as relational collaborators in high-stakes, community-led care.

Community engagement plays a critical role in anti-poaching efforts, yet existing mathematical models aimed at enhancing this engagement often overlook direct participation by community members as alternative patrollers. Unlike professional rangers, community members typically lack flexibility and experience, resulting in new challenges in optimizing patrol resource allocation. To address this gap, we propose a novel game-theoretic model for community-participated patrol, where a conservation agency strategically deploys both professional rangers and community members to safeguard wildlife against a best-responding poacher. In addition to a mixed-integer linear program formulation, we introduce a Two-Dimensional Binary Search algorithm and a novel Hybrid Waterfilling algorithm to efficiently solve the game in polynomial time. Through extensive experiments and a detailed case study focused on a protected tiger habitat in Northeast China, we demonstrate the effectiveness of our algorithms and the practical applicability of our model.

We consider the Coalition Structure Learning (CSL) problem in multi-agent systems, motivated by the existence of coalitions in many real-world systems, e.g., trading platforms and auction systems. In this problem, there is a hidden coalition structure within a set of n agents, which affects the behavior of the agents in games. Our goal is to actively design a sequence of games for the agents to play, such that observations in these games can be used to learn the hidden coalition structure. In particular, we consider the setting where in each round, we design and present a game together with a strategy profile to the agents, and receive a multiple-bit observation -- for each agent, we observe whether or not they would like to deviate from the specified strategy. We show that we can learn the coalition structure in O(log n) rounds if we are allowed to design any normal-form game, matching the information-theoretical lower bound. For practicality, we extend the result to settings where we can only choose games of a specific format, and design algorithms to learn the coalition structure in these settings. For most settings, our complexity matches the theoretical lower bound up to a constant factor.

Large Language Models (LLMs) have demonstrated strong reasoning capabilities in solving complex problems. However, current approaches primarily enhance reasoning through the elaboration of thoughts while neglecting the diversity of reasoning types. LLMs typically employ deductive reasoning, proceeding step-by-step from given conditions, which limits their exploration during problem-solving. Our analysis reveals that certain problems are exclusively solvable through specific reasoning strategies like inductive, abductive, or analogical reasoning. However, incorporating diverse reasoning approaches presents two key challenges: identifying the appropriate reasoning type for each problem and exploiting this approach during problem-solving. Therefore, we propose the TypedThinker that predicts suitable reasoning types based on the problem and their previous effectiveness and provides relevant demonstrations to guide LLMs in applying these strategies. Experimental results show significant improvements across multiple benchmarks, with performance gains of 3.4\% for Mistral 7B, 6.5\% for LLaMA3 8B, and 7\% for Qwen 2 7B on logical and mathematical reasoning tasks. TypedThinker enhances LLM reasoning without requiring knowledge distillation from larger models. It can be integrated into more advanced systems like GPT-4o or specialized models like MetaMath to diversify their reasoning approaches and improve their problem-solving capabilities.

Recent advances in reinforcement learning (RL) have predominantly leveraged neural network policies for decision-making, yet these models often lack interpretability, posing challenges for stakeholder comprehension and trust. Concept bottleneck models offer an interpretable alternative by integrating human-understandable concepts into policies. However, prior work assumes that concept annotations are readily available during training. For RL, this requirement poses a significant limitation: it necessitates continuous real-time concept annotation, which either places an impractical burden on human annotators or incurs substantial costs in API queries and inference time when employing automated labeling methods. To overcome this limitation, we introduce a novel training scheme that enables RL agents to efficiently learn a concept-based policy by only querying annotators to label a small set of data. Our algorithm, LICORICE, involves three main contributions: interleaving concept learning and RL training, using an ensemble to actively select informative data points for labeling, and decorrelating the concept data. We show how LICORICE reduces human labeling efforts to 500 or fewer concept labels in three environments, and 5000 or fewer in two more complex environments, all at no cost to performance. We also explore the use of VLMs as automated concept annotators, finding them effective in some cases but imperfect in others. Our work significantly reduces the annotation burden for interpretable RL, making it more practical for real-world applications that necessitate transparency. Our code is released.

Designing effective reward functions in multi-agent reinforcement learning (MARL) is a significant challenge, often leading to suboptimal or misaligned behaviors in complex, coordinated environments. We introduce Multi-agent Reinforcement Learning from Multi-phase Human Feedback of Mixed Quality ($\text{M}^3\text{HF}$), a novel framework that integrates multi-phase human feedback of mixed quality into the MARL training process. By involving humans with diverse expertise levels to provide iterative guidance, $\text{M}^3\text{HF}$ leverages both expert and non-expert feedback to continuously refine agents' policies. During training, we strategically pause agent learning for human evaluation, parse feedback using large language models to assign it appropriately and update reward functions through predefined templates and adaptive weights by using weight decay and performance-based adjustments. Our approach enables the integration of nuanced human insights across various levels of quality, enhancing the interpretability and robustness of multi-agent cooperation. Empirical results in challenging environments demonstrate that $\text{M}^3\text{HF}$ significantly outperforms state-of-the-art methods, effectively addressing the complexities of reward design in MARL and enabling broader human participation in the training process.

Mian Zhang, Xianjun Yang, Xinlu Zhang, Travis Labrum, Jamie C. Chiu, Shaun M. Eack, Fei Fang, William Yang Wang, Zhiyu Chen. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Learning Nash equilibrium (NE) in complex zero-sum games with multi-agent reinforcement learning (MARL) can be extremely computationally expensive. Curriculum learning is an effective way to accelerate learning, but an under-explored dimension for generating a curriculum is the difficulty-to-learn of the subgames –games induced by starting from a specific state. In this work, we present a novel subgame curriculum learning framework for zero-sum games. It adopts an adaptive initial state distribution by resetting agents to some previously visited states where they can quickly learn to improve performance. Building upon this framework, we derive a subgame selection metric that approximates the squared distance to NE values and further adopt a particle-based state sampler for subgame generation. Integrating these techniques leads to our new algorithm, Subgame Automatic Curriculum Learning (SACL), which is a realization of the subgame curriculum learning framework. SACL can be combined with any MARL algorithm such as MAPPO. Experiments in the particle-world environment and Google Research Football environment show SACL produces much stronger policies than baselines. In the challenging hide-and-seek quadrant environment, SACL produces all four emergent stages and uses only half the samples of MAPPO with self-play. The project website is at https://sites.google.com/view/sacl-neurips.

Environmental conservation organizations routinely monitor news content on conservation in protected areas to maintain situational awareness of developments that can have an environmental impact. Existing automated media monitoring systems require large amounts of data labeled by domain experts, which is only feasible at scale for high-resource languages like English. However, such tools are most needed in the global south where the news of interest is mainly in local low-resource languages, and far fewer experts are available to annotate datasets on a sustainable basis. In this paper, we propose NewsSerow, a method to automatically recognize environmental conservation content in low-resource languages. NewsSerow is a pipeline of summarization, in-context few-shot classification, and self-reflection using large language models (LLMs). Using at most 10 demonstration example news articles in Nepali, NewsSerow significantly outperforms other few-shot methods and can achieve comparable performance with models fully fine-tuned using thousands of examples. With NewsSerow, Organization X has been able to deploy the media monitoring tool in Nepal, significantly reducing their operational burden, and ensuring that AI tools for conservation actually reach the communities that need them the most. NewsSerow has also been deployed for countries with other languages like Colombia.

Coalitions naturally exist in many real-world systems involving multiple decision makers such as ridesharing, security, and online ad auctions, but the coalition structure among the agents is often unknown. We propose and study an important yet previously overseen problem -- Coalition Structure Learning (CSL), where we aim to carefully design a series of games for the agents and infer the underlying coalition structure by observing their interactions in those games. We establish a lower bound on the sample complexity -- defined as the number of games needed to learn the structure -- of any algorithms for CSL and propose the Iterative Grouping (IG) algorithm for designing normal-form games to achieve the lower bound. We show that IG can be extended to other succinct games such as congestion games and graphical games. Moreover, we solve CSL in a more restrictive and practical setting: auctions. We show a variant of IG to solve CSL in the auction setting even if we cannot design the bidder valuations. Finally, we conduct experiments to evaluate IG in the auction setting and the results align with our theoretical analysis.

Ruiyi Wang, Stephanie Milani, Jamie C. Chiu, Jiayin Zhi, Shaun M. Eack, Travis Labrum, Samuel M Murphy, Nev Jones, Kate V Hardy, Hong Shen, Fei Fang, Zhiyu Chen. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Agents built with large language models (LLMs) have shown great potential across a wide range of domains. However, in complex decision-making tasks, pure LLM-based agents tend to exhibit intrinsic bias in their choice of actions, which is inherited from the model’s training data and results in suboptimal performance. To develop strategic language agents, i.e., agents that generate flexible language actions and possess strong decision-making abilities, we propose a novel framework that powers LLM-based agents with reinforcement learning (RL). We consider Werewolf, a popular social deduction game, as a challenging testbed that emphasizes versatile communication and strategic gameplay. To mitigate the intrinsic bias in language actions, our agents use an LLM to perform deductive reasoning and generate a diverse set of action candidates. Then an RL policy trained to optimize the decision-making ability chooses an action from the candidates to play in the game. Extensive experiments show that our agents overcome the intrinsic bias and outperform existing LLM-based agents in the Werewolf game. We also conduct human-agent experiments and find that our agents achieve human-level performance and demonstrate strong strategic play.

Restless multi-armed bandits (RMAB) extend multi-armed bandits so arm pulls impact future arm states. Despite the success of RMABs, a key limiting assumption is the separability of rewards into a sum across arms. We address this deficiency by proposing restless-multi-armed bandit with global rewards (RMAB-G), a generalization of RMABs to global non-separable rewards. To solve RMAB-G, we develop the Linear-Whittle and Shapley-Whittle indices, which extend Whittle indices from RMABs to RMAB-Gs. We prove approximation bounds which demonstrate how Linear and Shapley-Whittle indices fail for non-linear rewards. To overcome this limitation, we propose two sets of adaptive policies: the first computes indices iteratively and the second combines indices with Monte-Carlo Tree Search (MCTS). Empirically, we demonstrate that adaptive policies outperform both pre-computed index policies and baselines in synthetic and real-world food rescue datasets.

We study a multi-agent imitation learning (MAIL) problem where we take the perspective of a learner attempting to *coordinate* a group of agents based on demonstrations of an expert doing so. Most prior work in MAIL essentially reduces the problem to matching the behavior of the expert *within* the support of the demonstrations. While doing so is sufficient to drive the *value gap* between the learner and the expert to zero under the assumption that agents are non-strategic, it does not guarantee robustness to deviations by strategic agents. Intuitively, this is because strategic deviations can depend on a counterfactual quantity: the coordinator's recommendations outside of the state distribution their recommendations induce. In response, we initiate the study of an alternative objective for MAIL in Markov Games we term the *regret gap* that explicitly accounts for potential deviations by agents in the group. We first perform an in-depth exploration of the relationship between the value and regret gaps. First, we show that while the value gap can be efficiently minimized via a direct extension of single-agent IL algorithms, even *value equivalence* can lead to an arbitrarily large regret gap. This implies that achieving regret equivalence is harder than achieving value equivalence in MAIL. We then provide a pair of efficient reductions to no-regret online convex optimization that are capable of minimizing the regret gap *(a)* under a coverage assumption on the expert (MALICE) or *(b)* with access to a queryable expert (BLADES).

Food waste and food insecurity are two problems that co-exist worldwide. A major force to combat food waste and insecurity, food rescue platforms (FRP) match food donations to low-resource communities. Since they rely on external volunteers to deliver the food, communicating rescue task difficulty to volunteers is very important for volunteer engagement and retention. We develop a hybrid model with tabular and natural language data to predict the difficulty of a given rescue trip, which significantly outperforms baselines in identifying easy and hard rescues. Furthermore, using storyboards, we conducted interviews with different stakeholders to understand their perspectives on how to integrate such predictions into volunteers' workflow. Motivated by our findings, we developed three explanation methods to generate interpretable insights for volunteers to better understand the predictions. The results from this study are in the process of being adopted at Food Rescue Hero, a large FRP serving over 25 cities across the United States.

Non-governmental organizations for environmental conservation have a significant interest in monitoring conservation-related media and getting timely updates about infrastructure construction projects as they may cause massive impact to key conservation areas. Such monitoring, however, is difficult and time-consuming. We introduce NewsPanda, a toolkit which automatically detects and analyzes online articles related to environmental conservation and infrastructure construction. We fine-tune a BERT-based model using active learning methods and noise correction algorithms to identify articles that are relevant to conservation and infrastructure construction. For the identified articles, we perform further analysis, extracting keywords and finding potentially related sources. NewsPanda has been successfully deployed by the World Wide Fund for Nature teams in the UK, India, and Nepal since February 2022. It currently monitors over 80,000 websites and 1,074 conservation sites across India and Nepal, saving more than 30 hours of human efforts weekly. We have now scaled it up to cover 60,000 conservation sites globally.

Function approximation (FA) has been a critical component in solving large zero-sum games. Yet, little attention has been given towards FA in solving general-sum extensive-form games, despite them being widely regarded as being computationally more challenging than their fully competitive or cooperative counterparts. A key challenge is that for many equilibria in general-sum games, no simple analogue to the state value function used in Markov Decision Processes and zero-sum games exists. In this paper, we propose learning the Enforceable Payoff Frontier (EPF)---a generalization of the state value function for general-sum games. We approximate the optimal Stackelberg extensive-form correlated equilibrium by representing EPFs with neural networks and training them by using appropriate backup operations and loss functions. This is the first method that applies FA to the Stackelberg setting, allowing us to scale to much larger games while still enjoying performance guarantees based on FA error. Additionally, our proposed method guarantees incentive compatibility and is easy to evaluate without having to depend on self-play or approximate best-response oracles.

We aim to understand how people assess human likeness in navigation produced by people and artificially intelligent (AI) agents in a video game. To this end, we propose a novel AI agent with the goal of generating more human-like behavior. We collect hundreds of crowd-sourced assessments comparing the human-likeness of navigation behavior generated by our agent and baseline AI agents with human-generated behavior. Our proposed agent passes a Turing Test, while the baseline agents do not. By passing a Turing Test, we mean that human judges could not quantitatively distinguish between videos of a person and an AI agent navigating. To understand what people believe constitutes human-like navigation, we extensively analyze the justifications of these assessments. This work provides insights into the characteristics that people consider human-like in the context of goal-directed video game navigation, which is a key step for further improving human interactions with AI agents.

Action advising is a knowledge transfer technique for reinforcement learning based on the teacher-student paradigm. An expert teacher provides advice to a student during training in order to improve the student's sample efficiency and policy performance. Such advice is commonly given in the form of state-action pairs. However, it makes it difficult for the student to reason with and apply to novel states. We introduce Explainable Action Advising, in which the teacher provides action advice as well as associated explanations indicating why the action was chosen. This allows the student to self-reflect on what it has learned, enabling advice generalization and leading to improved sample efficiency and learning performance - even in environments where the teacher is sub-optimal. We empirically show that our framework is effective in both single-agent and multi-agent scenarios, yielding improved policy returns and convergence rates when compared to state-of-the-art methods.

The assignment of papers to reviewers is a crucial part of the peer review processes of large publication venues, where organizers (e.g., conference program chairs) rely on algorithms to perform automated paper assignment. As such, a major challenge for the organizers of these processes is to specify paper assignment algorithms that find appropriate assignments with respect to various desiderata. Although the main objective when choosing a good paper assignment is to maximize the expertise of each reviewer for their assigned papers, several other considerations make introducing randomization into the paper assignment desirable: robustness to malicious behavior, the ability to evaluate alternative paper assignments, reviewer diversity, and reviewer anonymity. However, it is unclear in what way one should randomize the paper assignment in order to best satisfy all of these considerations simultaneously. In this work, we present a practical, one-size-fits-all method for randomized paper assignment intended to perform well across different motivations for randomness. We show theoretically and experimentally that our method outperforms currently-deployed methods for randomized paper assignment on several intuitive randomness metrics, demonstrating that the randomized assignments produced by our method are general-purpose.

Given a tensor that captures temporal data, such as (user, item, time), the way that we set the granularity of the “time” mode can make or break our analysis of the data. If we set the granularity to be extremely fine, we end up with a very sparse and high-rank tensor which is essentially incompatible with what virtually all tensor decomposition models expect, i.e., tensors with low-rank structure, which can be expressed in some form of factorization. Traditionally, this problem has been avoided by setting the granularity of the “time” to a “reasonable” aggregation (say hourly or daily intervals), an approach which has certainly served tensor analysis of temporal methods well so far. However, such an approach requires tedious trial- and-error experimentation across a number of such fixed aggregations, where typically the one that provides the most sensible results is retained, and furthermore it is arbitrary, since the optimal aggregation over time need not necessarily be uniform. In our work, we directly tackle this problem. We introduce Harvester, the first principled factorization-based approach which seeks to identify the best temporal granularity of a given tensor. Unlike existing methods which follow a greedy approach, Harvester leverages multiple aggregated views of the tensor, and a carefully-designed optimization problem, in order to uncover an aggregation of a tensor which has a “good” structure for factor analysis or a downstream task. We extensively evaluate Harvester on synthetic and real data, and demonstrate that it consistently produces tensors of very high quality, compared to the state-of-the-art, across the board for a number of different popular quality measures that have been used by the community.

In conference peer review, reviewers are often asked to provide “bids” on each submitted paper that express their interest in reviewing that paper. A paper assignment algorithm then uses these bids (along with other data) to compute a high-quality assignment of reviewers to papers. However, this process has been exploited by malicious reviewers who strategically bid in order to unethically manipulate the paper assignment, crucially undermining the peer review process. For example, these reviewers may aim to get assigned to a friend’s paper as part of a quid-pro-quo deal. A critical impediment towards creating and evaluating methods to mitigate this issue is the lack of any publicly-available data on malicious paper bidding. In this work, we collect and publicly release a novel dataset to fill this gap, collected from a mock conference activity where participants were instructed to bid either honestly or maliciously. We further provide a descriptive analysis of the bidding behavior, including our categorization of different strategies employed by participants. Finally, we evaluate the ability of each strategy to manipulate the assignment, and also evaluate the performance of some simple algorithms meant to detect malicious bidding. The performance of these detection algorithms can be taken as a baseline for future research on detecting malicious bidding.

Food waste and insecurity are two societal challenges that coexist in many parts of the world. A prominent force to combat these issues, food rescue platforms match food donations to organizations that serve underprivileged communities, and then rely on external volunteers to transport the food. Previous work has developed machine learning models for food rescue volunteer engagement. However, having long worked with domain practitioners to deploy AI tools to help with food rescues, we understand that there are four main pain points that keep such a machine learning model from being actually useful in practice: small data, data collected only under the default intervention, unmodeled objectives due to communication gap, and unforeseen consequences of the intervention. In this paper, we introduce bandit data-driven optimization which not only helps address these pain points in food rescue, but also is applicable to other nonprofit domains that share similar challenges. Bandit data-driven optimization combines the advantages of online bandit learning and offline predictive analytics in an integrated framework. We propose PROOF, a novel algorithm for this framework and formally prove that it has no-regret. We show that PROOF performs better than existing baseline on food rescue volunteer recommendation.

Many scientific conferences employ a two-phase paper review process, where some papers are assigned additional reviewers after the initial reviews are submitted. Many conferences also design and run experiments on their paper review process, where some papers are assigned reviewers who provide reviews under an experimental condition. In this paper, we consider the question: how should reviewers be divided between phases or conditions in order to maximize total assignment similarity? We make several contributions towards answering this question. First, we prove that when the set of papers requiring additional review is unknown, a simplified variant of this problem is NP-hard. Second, we empirically show that across several datasets pertaining to real conference data, dividing reviewers between phases/conditions uniformly at random allows an assignment that is nearly as good as the oracle optimal assignment. This uniformly random choice is practical for both the two-phase and conference experiment design settings. Third, we provide explanations of this phenomenon by providing theoretical bounds on the suboptimality of this random strategy under certain natural conditions. From these easily-interpretable conditions, we provide actionable insights to conference program chairs about whether a random reviewer split is suitable for their conference.

Robust Reinforcement Learning (RL) focuses on improving performances under model errors or adversarial attacks, which facilitates the real-life deployment of RL agents. Robust Adversarial Reinforcement Learning (RARL) is one of the most popular frameworks for robust RL. However, most of the existing literature models RARL as a zero-sum simultaneous game with Nash equilibrium as the solution concept, which could overlook the sequential nature of RL deployments, produce overly conservative agents, and induce training instability. In this paper, we introduce a novel hierarchical formulation of robust RL -- a general-sum Stackelberg game model called RRL-Stack -- to formalize the sequential nature and provide extra flexibility for robust training. We develop the Stackelberg Policy Gradient algorithm to solve RRL-Stack, leveraging the Stackelberg learning dynamics by considering the adversary's response. Our method generates challenging yet solvable adversarial environments which benefit RL agents' robust learning. Our algorithm demonstrates better training stability and robustness against different testing conditions in the single-agent robotics control and multi-agent highway merging tasks.

Preventing poaching through ranger patrols protects endangered wildlife, directly contributing to the UN Sustainable Development Goal 15 of life on land. Combinatorial bandits have been used to allocate limited patrol resources, but existing approaches overlook the fact that each location is home to multiple species in varying proportions, so a patrol benefits each species to differing degrees. When some species are more vulnerable, we ought to offer more protection to these animals; unfortunately, existing combinatorial bandit approaches do not offer a way to prioritize important species. To bridge this gap, (1) We propose a novel combinatorial bandit objective that trades off between reward maximization and also accounts for prioritization over species, which we call ranked prioritization. We show this objective can be expressed as a weighted linear sum of Lipschitz-continuous reward functions. (2) We provide RankedCUCB, an algorithm to select combinatorial actions that optimize our prioritization-based objective, and prove that it achieves asymptotic no-regret. (3) We demonstrate empirically that RankedCUCB leads to up to 38% improvement in outcomes for endangered species using real-world wildlife conservation data. Along with adapting to other challenges such as preventing illegal logging and overfishing, our no-regret algorithm addresses the general combinatorial bandit problem with a weighted linear objective.

Curriculum Reinforcement Learning (CRL) aims to create a sequence of tasks, starting from easy ones and gradually learning towards difficult tasks. In this work, we focus on the idea of framing CRL as interpolations between a source (auxiliary) and a target task distribution. Although existing studies have shown the great potential of this idea, it remains unclear how to formally quantify and generate the movement between task distributions. Inspired by the insights from gradual domain adaptation in semi-supervised learning, we create a natural curriculum by breaking down the potentially large task distributional shift in CRL into smaller shifts. We propose GRADIENT which formulates CRL as an optimal transport problem with a tailored distance metric between tasks. Specifically, we generate a sequence of task distributions as a geodesic interpolation between the source and target distributions, which are actually the Wasserstein barycenter. Different from many existing methods, our algorithm considers a task-dependent contextual distance metric and is capable of handling nonparametric distributions in both continuous and discrete context settings. In addition, we theoretically show that GRADIENT enables smooth transfer between subsequent stages in the curriculum under certain conditions. We conduct extensive experiments in locomotion and manipulation tasks and show that our proposed GRADIENT achieves higher performance than baselines in terms of learning efficiency and asymptotic performance.

Optimizing strategic decisions (a.k.a. computing equilibrium) is key to the success of many non-cooperative multi-agent applications. However, in many real-world situations, we may face the exact opposite of this game-theoretic problem --- instead of prescribing equilibrium of a given game, we may directly observe the agents' equilibrium behaviors but want to infer the underlying parameters of an unknown game. This research question, also known as inverse game theory, has been studied in multiple recent works in the context of Stackelberg games. Unfortunately, existing works exhibit quite negative results, showing statistical hardness and computational hardness, assuming follower's perfectly rational behaviors. Our work relaxes the perfect rationality agent assumption to the classic quantal response model, a more realistic behavior model of bounded rationality. Interestingly, we show that the smooth property brought by such bounded rationality model actually leads to provably more efficient learning of the follower utility parameters in general Stackelberg games. Systematic empirical experiments on synthesized games confirm our theoretical results and further suggest its robustness beyond the strict quantal response model.

As a challenging multi-player card game, DouDizhu has recently drawn much attention for analyzing competition and collaboration in imperfect-information games. In this paper, we propose PerfectDou, a state-of-the-art Doudizhu AI system that summits the game, in an actor-critic framework with a proposed technique named perfect information distillation.In detail, we adopt a perfect-training-imperfection-execution framework that allows the agents to utilize the global information to guide the training of the policies as if it is a perfect information game and the trained policies can be used to play the imperfect information game during the actual gameplay. Correspondingly, we characterize card and game features for DouDizhu to represent the perfect and imperfect information. To train our system, we adopt proximal policy optimization with generalized advantage estimation in a parallel training paradigm. In experiments we show how and why PerfectDou beats all existing programs, and achieves state-of-the-art performance.

Energy harvesting (EH)-powered sensor nodes can achieve theoretically unlimited lifetime by scavenging energy from ambient power sources, such as radio-frequency (RF) and kinetic energy. The nodes can collect and transmit data wirelessly with the harvested energy. However, the transmission between two sensor nodes is successful only when both nodes have enough energy at the same time. While the receiver can be actively listening, it may deplete the energy long before the sender has accumulated enough energy. Thus, given the scarce, unpredictable, and unevenly distributed energy among sensor nodes, it is challenging to ensure efficient data transmission between them. To address this challenge, we propose a sensor node architecture with multiple radios, each with different energy consumption on the sender and receiver. A node can be put into sleep when charged up and wakes up for communication when it infers that both nodes have enough energy based on its observations. What is more, two nodes can cooperatively and dynamically select different radios according to the stored energy and historical information to maximize the data throughput. To achieve cooperative communication adaptively, the communication procedure is modeled as a cooperative Markov game with partial observability on each node, and multiagent reinforcement learning (MARL) is employed to achieve the best results. Experimental results on hardware prototype and by simulation show that the proposed approaches achieve up to 89.1% of the optimal throughput and significantly outperform other online algorithms.

Wildlife poaching is threatening key species that play important roles in the ecosystem. With historical ranger patrol records, it is possible to provide data-driven predictions of poaching threats and plan patrols to combat poaching. However, the patrollers often patrol in a multimodal way, which combines driving and walking. It is a tedious task for the domain experts to manually plan such a patrol and as a result, the planned patrol routes are often far from optimal. In this paper, we propose a data-driven approach for multimodal patrol planning. We first use machine learning models to predict the poaching threats and then use a novel mixed-integer linear programming-based algorithm to plan the patrol route. In a field test focusing on the machine learning prediction result at Jilin Huangnihe National Nature Reserve (HNHR) in December 2019, the rangers found 42 snares, which is significantly higher than the historical record. Our offline experiments show that the resulting multimodal patrol routes can improve the efficiency of patrol and thus they can serve as the basis for future deployment in the field.

Current work in explainable reinforcement learning generally produces policies in the form of a decision tree over the state space. Such policies can be used for formal safety verification, agent behavior prediction, and manual inspection of important features. However, existing approaches fit a decision tree after training or use a custom learning procedure which is not compatible with new learning techniques, such as those which use neural networks. To address this limitation, we propose a novel Markov Decision Process (MDP) type for learning decision tree policies: Iterative Bounding MDPs (IBMDPs). An IBMDP is constructed around a base MDP so each IBMDP policy is guaranteed to correspond to a decision tree policy for the base MDP when using a method-agnostic masking procedure. Because of this decision tree equivalence, any function approximator can be used during training, including a neural network, while yielding a decision tree policy for the base MDP. We present the required masking procedure as well as a modified value update step which allows IBMDPs to be solved using existing algorithms. We apply this procedure to produce IBMDP variants of recent reinforcement learning methods. We empirically show the benefits of our approach by solving IBMDPs to produce decision tree policies for the base MDPs.

Conservation efforts in green security domains to protect wildlife and forests are constrained by the limited availability of defenders (i.e., patrollers), who must patrol vast areas to protect from attackers (e.g., poachers or illegal loggers). Defenders must choose how much time to spend in each region of the protected area, balancing exploration of infrequently visited regions and exploitation of known hotspots. We formulate the problem as a stochastic multi-armed bandit, where each action represents a patrol strategy, enabling us to guarantee the rate of convergence of the patrolling policy. However, a naive bandit approach would compromise short-term performance for long-term optimality, resulting in animals poached and forests destroyed. To speed up performance, we leverage smoothness in the reward function and decomposability of actions. We show a synergy between Lipschitz-continuity and decomposition as each aids the convergence of the other. In doing so, we bridge the gap between combinatorial and Lipschitz bandits, presenting a no-regret approach that tightens existing guarantees while optimizing for short-term performance. We demonstrate that our algorithm, LIZARD, improves performance on real-world poaching data from Cambodia.

Food banks provide communities and organizations with food for those in need. One challenge they face is properly estimating the resources needed to fulfill orders. Estimating the number of shipping pallets needed for each order is an important step in allocating these resources, and coupled with limited data, provides a challenging mental task which the food bank staff grapple with on a daily basis. We provide an algorithm to estimate the number of pallets needed for an order based on the quantity of products, the known products-per-tier, and tiers-per-pallet values, as well as a scheme for testing this algorithm with limited data from the food bank. The algorithm aids in resource allocation by reducing uncertainty in the number of pallets needed.

One practical requirement in solving dynamic games is to ensure that the players play well from any decision point onward. To satisfy this requirement, existing efforts focus on equilibrium refinement, but the scalability and applicability of existing techniques are limited. In this paper, we propose Temporal-Induced Self-Play (TISP), a novel reinforcement learning-based framework to find strategies with decent performances from any decision point onward. TISP uses belief-space representation, backward induction, policy learning, and non-parametric approximation. Building upon TISP, we design a policy-gradient-based algorithm TISP-PG. We prove that TISP-based algorithms can find approximate Perfect Bayesian Equilibrium in zero-sum one-sided stochastic Bayesian games with finite horizon. We test TISP-based algorithms in various games, including finitely repeated security games and a grid-world game. The results show that TISP-PG is more scalable than existing mathematical programming-based methods and significantly outperforms other learning-based methods.

Green security domains feature defenders who plan patrols in the face of uncertainty about the adversarial behavior of poachers, illegal loggers, and illegal fishers. Importantly, the deterrence effect of patrols on adversaries’ future behavior makes patrol planning a sequential decision-making problem. Therefore, we focus on robust sequential patrol planning for green security following the minimax regret criterion, which has not been considered in the literature. We formulate the problem as a game between the defender and nature who controls the parameter values of the adversarial behavior and design an algorithm MIRROR to find a robust policy. MIRROR uses two reinforcement learning–based oracles and solves a restricted game considering limited defender strategies and parameter values. We evaluate MIRROR on real-world poaching data.

The challenges of food waste and insecurity arise in wealthy and developing nations alike, impacting millions of livelihoods. The ongoing pandemic only exacerbates the problem. A major force to combat food waste and insecurity, food rescue (FR) organizations match food donations to the non-profits that serve low-resource communities. Since they rely on external volunteers to pick up and deliver the food, some FRs use web-based mobile applications to reach the right set of volunteers. In this paper, we propose the first machine learning based model to improve volunteer engagement in the food waste and security domain. We (1) develop a recommender system to send push notifications to the most likely volunteers for each given rescue, (2) leverage a mathematical programming based approach to diversify our recommendations, and (3) propose an online algorithm to dynamically select the volunteers to notify without the knowledge of future rescues. Our recommendation system improves the hit ratio from 44% achieved by the previous method to 73%. A pilot study of our method is scheduled to take place in the near future.

Motivated by real-world deployment of drones for conservation, this paper advances the state-of-the-art in security games with signaling. The well-known defender-attacker security games framework can help in planning for such strategic deployments of sensors and human patrollers, and warning signals to ward off adversaries. However, we show that defenders can suffer significant losses when ignoring real-world uncertainties despite carefully planned security game strategies with signaling. In fact, defenders may perform worse than forgoing drones completely in this case. We address this shortcoming by proposing a novel game model that integrates signaling and sensor uncertainty; perhaps surprisingly, we show that defenders can still perform well via a signaling strategy that exploits uncertain real-time information. For example, even in the presence of uncertainty, the defender still has an informational advantage in knowing that she has or has not actually detected the attacker; and she can design a signaling scheme to “mislead” the attacker who is uncertain as to whether he has been detected. We provide theoretical results, a novel algorithm, scale-up techniques, and experimental results from simulation based on our ongoing deployment of a conservation drone system in South Africa.

Cyber adversaries have increasingly leveraged social engineering attacks to breach large organizations and threaten the well-being of today's online users. One clever technique, the “watering hole” attack, compromises a legitimate website to execute drive-by download attacks by redirecting users to another malicious domain. We introduce a game-theoretic model that captures the salient aspects for an organization protecting itself from a watering hole attack by altering the environment information in web traffic so as to deceive the attackers. Our main contributions are (1) a novel Social Engineering Deception (SED) game model that features a continuous action set for the attacker, (2) an in-depth analysis of the SED model to identify computationally feasible real-world cases, and (3) the CyberTWEAK algorithm which solves for the optimal protection policy. To illustrate the potential use of our framework, we built a browser extension based on our algorithms which is now publicly available online. The CyberTWEAK extension will be vital to the continued development and deployment of countermeasures for social engineering.

Food waste and food insecurity are two challenges that coexist in many communities. To mitigate the problem, food rescue platforms match excess food with the communities in need, and leverage external volunteers to transport the food. However, the external volunteers bring significant uncertainty to the food rescue operation. We work with a large food rescue organization to predict the uncertainty and furthermore to find ways to reduce the human dispatcher's workload and the redundant notifications sent to volunteers. We make two main contributions. (1) We train a stacking model which predicts whether a rescue will be claimed with high precision and AUC. This model can help the dispatcher better plan for backup options and alleviate their uncertainty. (2) We develop a data-driven optimization algorithm to compute the optimal intervention and notification scheme. The algorithm uses a novel counterfactual data generation approach and the branch and bound framework. Our result reduces the number of notifications and interventions required in the food rescue operation. We are working with the organization to deploy our results in the near future.

Although security games have attracted intensive research attention over the past years, few existing works consider how information from local communities would affect the game. In this paper, we introduce a new player -- a strategic informant, who can observe and report upcoming attacks -- to the defender-attacker security game setting. Characterized by a private type, the informant has his utility structure that leads to his strategic behaviors. We model the game as a 3-player extensive-form game and propose a novel solution concept of Strong Stackelberg-perfect Bayesian equilibrium. To compute the optimal defender strategy, we first show that although the informant can have infinitely many types in general, the optimal defense plan can only include a finite (exponential) number of different patrol strategies. We then prove that there exists a defense plan with only a linear number of patrol strategies that achieve the optimal defender's utility, which significantly reduces the computational burden and allows us to solve the game in polynomial time using linear programming. Finally, we conduct extensive experiments to show the effect of the strategic informant and demonstrate the effectiveness of our algorithm.

We consider three important challenges in conference peer review: (i) reviewers maliciously attempting to get assigned to certain papers to provide positive reviews, possibly as part of quid-pro-quo arrangements with the authors; (ii) "torpedo reviewing," where reviewers deliberately attempt to get assigned to certain papers that they dislike in order to reject them; (iii) reviewer de-anonymization on release of the similarities and the reviewer-assignment code. On the conceptual front, we identify connections between these three problems and present a framework that brings all these challenges under a common umbrella. We then present a (randomized) algorithm for reviewer assignment that can optimally solve the reviewer-assignment problem under any given constraints on the probability of assignment for any reviewer-paper pair. We further consider the problem of restricting the joint probability that certain suspect pairs of reviewers are assigned to certain papers, and show that this problem is NP-hard for arbitrary constraints on these joint probabilities but efficiently solvable for a practical special case. Finally, we experimentally evaluate our algorithms on datasets from past conferences, where we observe that they can limit the chance that any malicious reviewer gets assigned to their desired paper to 50% while producing assignments with over 90% of the total optimal similarity.

A central problem in machine learning and statistics is to model joint densities of random variables from data. Copulas are joint cumulative distribution functions with uniform marginal distributions and are used to capture interdependencies in isolation from marginals. Copulas are widely used within statistics, but have not gained traction in the context of modern deep learning. In this paper, we introduce ACNet, a novel differentiable neural network architecture that enforces structural properties and enables one to learn an important class of copulas--Archimedean Copulas. Unlike Generative Adversarial Networks, Variational Autoencoders, or Normalizing Flow methods, which learn either densities or the generative process directly, ACNet learns a generator of the copula, which implicitly defines the cumulative distribution function of a joint distribution. We give a probabilistic interpretation of the network parameters of ACNet and use this to derive a simple but efficient sampling algorithm for the learned copula. Our experiments show that ACNet is able to both approximate common Archimedean Copulas and generate new copulas which may provide better fits to data.

Strong Stackelberg equilibrium (SSE) is the standard solution concept of Stackelberg security games. As opposed to the weak Stackelberg equilibrium (WSE), the SSE assumes that the follower breaks ties in favor of the leader and this is widely acknowledged and justified by the assertion that the defender can often induce the attacker to choose a preferred action by making an infinitesimal adjustment to her strategy. Unfortunately, in security games with resource assignment constraints, the assertion might not be valid; it is possible that the defender cannot induce the desired outcome. As a result, many results claimed in the literature may be overly optimistic. To remedy, we first formally define the utility guarantee of a defender strategy and provide examples to show that the utility of SSE can be higher than its utility guarantee. Second, inspired by the analysis of leader’s payoff by Von Stengel and Zamir (2004), we provide the solution concept called the inducible Stackelberg equilibrium (ISE), which owns the highest utility guarantee and always exists. Third, we show the conditions when ISE coincides with SSE and the fact that in general case, SSE can be extremely worse with respect to utility guarantee. Moreover, introducing the ISE does not invalidate existing algorithmic results as the problem of computing an ISE polynomially reduces to that of computing an SSE. We also provide an algorithmic implementation for computing ISE, with which our experiments unveil the empirical advantage of the ISE over the SSE.

Despite the recent advances of deep reinforcement learning (DRL), agents trained by DRL tend to be brittle and sensitive to the training environment, especially in the multi-agent scenarios. In the multi-agent setting, a DRL agent’s policy can easily get stuck in a poor local optima w.r.t. its training partners – the learned policy may be only locally optimal to other agents’ current policies. In this paper, we focus on the problem of training robust DRL agents with continuous actions in the multi-agent learning setting so that the trained agents can still generalize when its opponents’ policies alter. To tackle this problem, we proposed a new algorithm, MiniMax Multi-agent Deep Deterministic Policy Gradient (M3DDPG) with the following contributions: (1) we introduce a minimax extension of the popular multi-agent deep deterministic policy gradient algorithm (MADDPG), for robust policy learning; (2) since the continuous action space leads to computational intractability in our minimax learning objective, we propose Multi-Agent Adversarial Learning (MAAL) to efficiently solve our proposed formulation. We empirically evaluate our M3DDPG algorithm in four mixed cooperative and competitive multi-agent environments and the agents trained by our method significantly outperforms existing baselines.

With the recent advances in solving large, zero-sum extensive form games, there is a growing interest in the inverse problem of inferring underlying game parameters given only access to agent actions. Although a recent work provides a powerful differentiable end-to-end learning frameworks which embed a game solver within a deep-learning framework, allowing unknown game parameters to be learned via backpropagation, this framework faces significant limitations when applied to boundedly rational human agents and large scale problems, leading to poor practicality. In this paper, we address these limitations and propose a framework that is applicable for more practical settings. First, seeking to learn the rationality of human agents in complex two-player zero-sum games, we draw upon well-known ideas in decision theory to obtain a concise and interpretable agent behavior model, and derive solvers and gradients for end-to-end learning. Second, to scale up to large, real-world scenarios, we propose an efficient first-order primal-dual method which exploits the structure of extensive-form games, yielding significantly faster computation for both game solving and gradient computation. When tested on randomly generated games, we report speedups of orders of magnitude over previous approaches. We also demonstrate the effectiveness of our model on both real-world one-player settings and synthetic data.

Green Security Games (GSGs) have been proposed and applied to optimize patrols conducted by law enforcement agencies in green security domains such as combating poaching, illegal logging and overfishing. However, real-time information such as footprints and agents’ subsequent actions upon receiving the information, e.g., rangers following the footprints to chase the poacher, have been neglected in previous work. To fill the gap, we first propose a new game model GSG-I which augments GSGs with sequential movement and the vital element of real-time information. Second, we design a novel deep reinforcement learning-based algorithm, DeDOL, to compute a patrolling strategy that adapts to the real-time information against a best-responding attacker. DeDOL is built upon the double oracle framework and the policy-space response oracle, solving a restricted game and iteratively adding best response strategies to it through training deep Q-networks. Exploring the game structure, DeDOL uses domain-specific heuristic strategies as initial strategies and constructs several local modes for efficient and parallelized training. To our knowledge, this is the first attempt to use Deep Q-Learning for security games.

Real-world problems often involve more than one decision makers, each with their own goals or preferences. While game theory is an established paradigm for reasoning strategic interactions between multiple decision-makers, its applicability in practice is often limited by the intractability of computing equilibria in large games, and the fact that the game parameters are sometimes unknown and the players are often not perfectly rational. On the other hand, machine learning and reinforcement learning have led to huge successes in various domains and can be leveraged to overcome the limitations of the game-theoretic analysis. In this paper, we introduce our work on integrating learning with computational game theory for addressing societal challenges such as security and sustainability.

Advanced persistent threats (APTs) are a particularly troubling challenge for software systems. The adversarial nature of the security domain, and APTs in particular, poses unresolved challenges to the design of self-* systems, such as how to defend against multiple types of attackers with different goals and capabilities. In this interaction, the observability of each side is an important and under-investigated issue in the self-* domain. We propose a model of APT defense that elevates observability as a first-class concern. We evaluate this model by showing how an informed approach that uses observability improves the defender's utility compared to a uniform random strategy, can enable robust planning through sensitivity analysis, and can inform observability-related architectural design decisions.

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.

As robots are increasingly endowed with social and communicative capabilities, they will interact with humans in more settings, both collaborative and competitive. We explore human-robot relationships in the context of a competitive Stackelberg Security Game. We vary humanoid robot expressive language (in the form of “encouraging” or “discouraging” verbal commentary) and measure the impact on participants' rationality, strategy prioritization, mood, and perceptions of the robot. We learn that a robot opponent that makes discouraging comments causes a human to play a game less rationally and to perceive the robot more negatively. We also contribute a simple open source Natural Language Processing framework for generating expressive sentences, which was used to generate the speech of our autonomous social robot.

The unrelenting threat of poaching has led to increased development of new technologies to combat it. One such example is the use of long wave thermal infrared cameras mounted on unmanned aerial vehicles (UAVs or drones) to spot poachers at night and report them to park rangers before they are able to harm animals. However, monitoring the live video stream from these conservation UAVs all night is an arduous task. Therefore, we build SPOT (Systematic POacher deTector), a novel application that augments conservation drones with the ability to automatically detect poachers and animals in near real time. SPOT illustrates the feasibility of building upon state-of-the-art AI techniques, such as Faster RCNN, to address the challenges of automatically detecting animals and poachers in infrared images. This paper reports (i) the design and architecture of SPOT, (ii) a series of efforts towards more robust and faster processing to make SPOT usable in the field and provide detections in near real time, and (iii) evaluation of SPOT based on both historical videos and a real-world test run by the end users in the field. The promising results from the test in the field have led to a plan for larger-scale deployment in a national park in Botswana. While SPOT is developed for conservation drones, its design and novel techniques have wider application for automated detection from UAV videos.

A wealth of algorithms centered around (integer) linear programming have been proposed to compute equilibrium strategies in security games with discrete states and actions. However, in practice many domains possess continuous state and action spaces. In this paper, we consider a continuous space security game model with infinite-size action sets for players and present a novel deep learning based approach to extend the existing toolkit for solving security games. Specifically, we present (i) OptGradFP, a novel and general algorithm that searches for the optimal defender strategy in a parameterized continuous search space, and can also be used to learn policies over multiple game states simultaneously; (ii) OptGradFP-NN, a convolutional neural network based implementation of OptGradFP for continuous space security games. We demonstrate the potential to predict good defender strategies via experiments and analysis of OptGradFP and OptGradFP-NN on discrete and continuous game settings.

Increases in poaching levels have led to the use of unmanned aerial vehicles (UAVs or drones) to count animals, locate animals in parks, and even find poachers. Finding poachers is often done at night through the use of long wave thermal infrared cameras mounted on these UAVs. Unfortunately, monitoring the live video stream from the conservation UAVs all night is an arduous task. In order to assist in this monitoring task, new techniques in computer vision have been developed. This work is based on a dataset which took approximately six months to label. However, further improvement in detection and future testing of autonomous flight require not only more labeled training data, but also an environment where algorithms can be safely tested. In order to meet both goals efficiently, we present AirSim-W, a simulation environment that has been designed specifically for the domain of wildlife conservation. This includes (i) creation of an African savanna environment in Unreal Engine, (ii) integration of a new thermal infrared model based on radiometry, (iii) API code expansions to follow objects of interest or fly in zig-zag patterns to generate simulated training data, and (iv) demonstrated detection improvement using simulated data generated by AirSim-W. With these additional simulation features, AirSim-W will be directly useful for wildlife conservation research.

Poaching continues to be a significant threat to the conservation of wildlife and the associated ecosystem. Estimating and predicting where the poachers have committed or would commit crimes is essential to more effective allocation of patrolling resources. The real-world data in this domain is often sparse, noisy and incomplete, consisting of a small number of positive data (poaching signs), a large number of negative data with label uncertainty, and an even larger number of unlabeled data. Fortunately, domain experts such as rangers can provide complementary information about poaching activity patterns. However, this kind of human knowledge has rarely been used in previous approaches.

Although recent work in AI has made great progress in solving large, zero-sum, extensive-form games, the underlying assumption in most past work is that the parameters of the game itself are known to the agents. This paper deals with the relatively under-explored but equally important "inverse" setting, where the parameters of the underlying game are not known to all agents, but must be learned through observations. We propose a differentiable, end-to-end learning framework for addressing this task. In particular, we consider a regularized version of the game, equivalent to a particular form of quantal response equilibrium, and develop 1) a primal-dual Newton method for finding such equilibrium points in both normal and extensive form games; and 2) a backpropagation method that lets us analytically compute gradients of all relevant game parameters through the solution itself. This ultimately lets us learn the game by training in an end-to-end fashion, effectively by integrating a "differentiable game solver" into the loop of larger deep network architectures. We demonstrate the effectiveness of the learning method in several settings including poker and security game tasks.

We study Stackelberg Security Games where the defender, in addition to allocating defensive resources to protect targets from the attacker, can strategically manipulate the attacker’s payoff under budget constraints in weighted L^p-norm form regarding the amount of change. For the case of weighted L^1-norm constraint, we present (i) a mixed integer linear program-based algorithm with approximation guarantee; (ii) a branch-and-bound based algorithm with improved efficiency achieved by effective pruning; (iii) a polynomial time approximation scheme for a special but practical class of problems. In addition, we show that problems under budget constraints in L^0 and weighted L^\infty-norm form can be solved in polynomial time.

The Stackelberg Security Game (SSG) model has been immensely influential in security research since it was introduced roughly a decade ago. Furthermore, deployed SSG-based applications are one of most successful examples of game theory applications in the real world. We present a broad survey of recent technical advances in SSG and related literature, and then look to the future by highlighting the new potential applications and open research problems in SSG.

The conservation of key wildlife species such as tigers and elephants are threatened by poaching activities. In many conservation areas, foot patrols are conducted to prevent poaching but they may not be well-planned to make the best use of the limited patrolling resources. While prior work has introduced PAWS (Protection Assistant for Wildlife Security) as a game-theoretic decision aid to design effective foot patrol strategies to protect wildlife, the patrol routes generated by PAWS may be difficult to follow in areas with complex terrain. Subsequent research has worked on the significant evolution of PAWS, from an emerging application to a regularly deployed software. A key advance of the deployed version of PAWS is that it incorporates the complex terrain information and generates a strategy consisting of easy-to-follow routes. In this demonstration, we provide 1) a video introducing the PAWS system; 2) an interactive visualization of the patrol routes generated by PAWS in an example area with complex terrain; and 3) a machine-human competition in designing patrol strategy given complex terrain and animal distribution.

Poaching is a serious threat to the conservation of key species and whole ecosystems. While conducting foot patrols is the most commonly used approach in many countries to prevent poaching, such patrols often do not make the best use of limited patrolling resources. To remedy this situation, prior work introduced a novel emerging application called PAWS (Protection Assistant for Wildlife Security); PAWS was proposed as a game-theoretic (“security games”) decision aid to optimize the use of patrolling resources.This paper reports on PAWS’s significant evolution from a proposed decision aid to a regularly deployed application, reporting on the lessons from the first tests in Africa in Spring 2014, through its continued evolution since then, to current regular use in Southeast Asia and plans for future worldwide deployment. In this process, we have worked closely with two NGOs (Panthera and Rimba) and incorporated extensive feedback from professional patrolling teams. We outline key technical advances that lead to PAWS’s regular deployment: (i) incorporating complex topographic features, e.g., ridgelines, in generating patrol routes; (ii) handling uncertainties in species distribution (game theoretic payoffs); (iii) ensuring scalability for patrolling large-scale conservation areas with fine-grained guidance; and (iv) handling complex patrol scheduling constraints.

Recent years have seen increasing interest in AI from outside the AI community. This is partly due to applications based on AI that have been used in real-world domains, for example, the successful deployment of game theory-based decision aids in security domains. This paper describes our teaching approach for introducing the AI concepts underlying security games to diverse audiences. We adapted a game-based research platform that served as a testbed for recent research advances in computational game theory into a set of interactive role-playing games. We guided learners in playing these games as part of our teaching strategy, which also included didactic instruction and interactive exercises on broader AI topics. We describe our experience in applying this teaching approach to diverse audiences, including students of an urban public high school, university undergraduates, and security domain experts who protect wildlife. We evaluate our approach based on results from the games and participant surveys.

Fish stocks around the world are in danger from illegal fishing. In collaboration with the U.S. Coast Guard (USCG), we work to defend fisheries from illegal fisherman (henceforth called Lanchas) in the U.S. Gulf of Mexico. We have developed the COmPASS (Conservative Online Patrol ASSistant) system to design USCG patrols against the Lanchas. In this application, we face a population of Lanchas with heterogeneous behavior who fish frequently. We have some data about these Lanchas, but not enough to fit a statistical model. Previous security patrol assistants have focused on counterterrorism in one-shot games where adversaries are assumed to be perfectly rational, and much less data about their behavior is available. COmPASS is novel because: (i) it emphasizes environmental crime; (ii) it is based on a repeated Stackelberg game; (iii) it allows for bounded rationality of the Lanchas and it offers a robust approach against the heterogeneity of the Lancha population; and (iv) it can learn from sparse Lancha data. We report the effectiveness of COmPASS in the Gulf in our numerical experiments based on real fish data. The COmPASS system is to be tested by USCG.

Among the many deployment areas of Stackelberg Security games, a major area involves games played out in space and time, which includes applications in multiple mobile defender resources protecting multiple mobile targets. Previous algorithms for such spatio-temporal security games fail to scale-up and little is known ofthe computational complexity properties of these problems.This paper provides a novel oracle-based algorithmic framework for a systematic study of different problem variants of computing optimal (minimax) strategies in spatio-temporal security games. Our framework enables efficient computation of a minimax strategy when the problem admits a polynomial-time oracle. Furthermore,for the cases in which efficient oracles are difficultto find, we propose approximations or prove hardness results.

In recent years, Stackelberg Security Games have been successfully applied to solve resource allocation and scheduling problems in several security domains. However, previous work has mostly assumed that the targets are stationary relative to the defender and the attacker, leading to discrete game models with finite numbers of pure strategies. This paper in contrast focuses on protecting mobile targets that leads to a continuous set of strategies for the players. The problem is motivated by several real-world domains including protecting ferries with escort boats and protecting refugee supply lines. Our contributions include: (i) A new game model for multiple mobile defender resources and moving targets with a discretized strategy space for the defender and a continuous strategy space for the attacker. (ii) An efficient linear-programming-based solution that uses a compact representation for the defender's mixed strategy, while accurately modeling the attacker's continuous strategy using a novel sub-interval analysis method. (iii) Discussion and analysis of multiple heuristic methods for equilibrium refinement to improve robustness of defender's mixed strategy. (iv) Discussion of approaches to sample actual defender schedules from the defender's mixed strategy. (iv) Detailed experimental analysis of our algorithms in the ferry protection domain.

Illegal extraction of forest resources is fought, in many developing countries, by patrols that try to make this activity less profitable, using the threat of confiscation. With a limited budget, officials will try to distribute the patrols throughout the forest intelligently, in order to most effectively limit extraction. Prior work in forest economics has formalized this as a Stackelberg game, one very different in character from the discrete Stackelberg problem settings previously studied in the multiagent literature. Specifically, the leader wishes to minimize the distance by which a profit-maximizing extractor will trespass into the forest---or to maximize the radius of the remaining ``pristine'' forest area. The follower's cost-benefit analysis of potential trespass distances is affected by the likelihood of being caught and suffering confiscation. In this paper, we give a near-optimal patrol allocation algorithm and a 1/2-approximation algorithm, the latter of which is more efficient and yields simpler, more practical patrol allocations. Our simulations indicate that these algorithms substantially outperform existing heuristic allocations.