Search

Achieving seamless coordination in cooperative games is a crucial challenge in artificial intelligence, particularly when players operate under incomplete information. A common strategy to mitigate this information asymmetry involves leveraging explicit communication. However, direct communication is not always feasible due to factors such as transmission loss. We explore how effective coordination can be achieved without verbal communication, relying solely on observing each other's actions. We demonstrate how an autonomous agent can learn to cooperate by interpreting its partner's actions, which are used to hint at its intents. Our approach involves developing an agent strategy by constructing deterministic finite automata for each possible action and integrating them into a non-Markovian finite-state transducer. This transducer represents a non-deterministic strategy for the agent that suggests actions to assist its partner during gameplay. Experimental results in a testbed called Gnomes at Night show that the learned no-communication coordination strategy achieves significantly higher success rates and requires fewer steps to complete the game compared to uncoordinated scenarios, performing almost as well as an oracle baseline with direct communication., Comment: This paper is currently under review at AAAI 2025

In reinforcement learning, conducting task composition by forming cohesive, executable sequences from multiple tasks remains challenging. However, the ability to (de)compose tasks is a linchpin in developing robotic systems capable of learning complex behaviors. Yet, compositional reinforcement learning is beset with difficulties, including the high dimensionality of the problem space, scarcity of rewards, and absence of system robustness after task composition. To surmount these challenges, we view task composition through the prism of category theory -- a mathematical discipline exploring structures and their compositional relationships. The categorical properties of Markov decision processes untangle complex tasks into manageable sub-tasks, allowing for strategical reduction of dimensionality, facilitating more tractable reward structures, and bolstering system robustness. Experimental results support the categorical theory of reinforcement learning by enabling skill reduction, reuse, and recycling when learning complex robotic arm tasks., Comment: ECAI 2024

Negotiation is useful for resolving conflicts in multi-agent systems. We explore autonomous negotiation in a setting where two self-interested rational agents sequentially trade items from a finite set of categories. Each agent has a utility function that depends on the amount of items it possesses in each category. The offering agent makes trade offers to improve its utility without knowing the responding agent's utility function, and the responding agent accepts offers that improve its utility. We present a comparison-based algorithm for the offering agent that generates offers through previous acceptance or rejection responses without extensive information sharing. The algorithm estimates the responding agent's gradient by leveraging the rationality assumption and rejected offers to prune the space of potential gradients. After the algorithm makes a finite number of consecutively rejected offers, the responding agent is at a near-optimal state, or the agents' preferences are closely aligned. Additionally, we facilitate negotiations with humans by representing natural language feedback as comparisons that can be integrated into the proposed algorithm. We compare the proposed algorithm against random search baselines in integer and fractional trading scenarios and show that it improves the societal benefit with fewer offers.

Sociotechnical requirements shape the governance of artificially intelligent (AI) systems. In an era where embodied AI technologies are rapidly reshaping various facets of contemporary society, their inherent dynamic adaptability presents a unique blend of opportunities and challenges. Traditional regulatory mechanisms, often designed for static -- or slower-paced -- technologies, find themselves at a crossroads when faced with the fluid and evolving nature of AI systems. Moreover, typical problems in AI, for example, the frequent opacity and unpredictability of the behaviour of the systems, add additional sociotechnical challenges. To address these interconnected issues, we introduce the concept of dynamic certification, an adaptive regulatory framework specifically crafted to keep pace with the continuous evolution of AI systems. The complexity of these challenges requires common progress in multiple domains: technical, socio-governmental, and regulatory. Our proposed transdisciplinary approach is designed to ensure the safe, ethical, and practical deployment of AI systems, aligning them bidirectionally with the real-world contexts in which they operate. By doing so, we aim to bridge the gap between rapid technological advancement and effective regulatory oversight, ensuring that AI systems not only achieve their intended goals but also adhere to ethical standards and societal values.

Robust partially observable Markov decision processes (robust POMDPs) extend classical POMDPs to handle additional uncertainty on the transition and observation probabilities via so-called uncertainty sets. Policies for robust POMDPs must not only be memory-based to account for partial observability but also robust against model uncertainty to account for the worst-case instances from the uncertainty sets. We propose the pessimistic iterative planning (PIP) framework, which finds robust memory-based policies for robust POMDPs. PIP alternates between two main steps: (1) selecting an adversarial (non-robust) POMDP via worst-case probability instances from the uncertainty sets; and (2) computing a finite-state controller (FSC) for this adversarial POMDP. We evaluate the performance of this FSC on the original robust POMDP and use this evaluation in step (1) to select the next adversarial POMDP. Within PIP, we propose the rFSCNet algorithm. In each iteration, rFSCNet finds an FSC through a recurrent neural network trained using supervision policies optimized for the adversarial POMDP. The empirical evaluation in four benchmark environments showcases improved robustness against a baseline method in an ablation study and competitive performance compared to a state-of-the-art robust POMDP solver.

Recent years have seen increased awareness of the potential significant impacts of computing technologies, both positive and negative. This whitepaper explores how to address possible harmful consequences of computing technologies that might be difficult to anticipate, and thereby mitigate or address. It starts from the assumption that very few harms due to technology are intentional or deliberate; rather, the vast majority result from failure to recognize and respond to them prior to deployment. Nonetheless, there are concrete steps that can be taken to address the difficult problem of anticipating and responding to potential harms from new technologies.

Games with incomplete preferences are an important model for studying rational decision-making in scenarios where players face incomplete information about their preferences and must contend with incomparable outcomes. We study the problem of computing Nash equilibrium in a subclass of two-player games played on graphs where each player seeks to maximally satisfy their (possibly incomplete) preferences over a set of temporal goals. We characterize the Nash equilibrium and prove its existence in scenarios where player preferences are fully aligned, partially aligned, and completely opposite, in terms of the well-known solution concepts of sure winning and Pareto efficiency. When preferences are partially aligned, we derive conditions under which a player needs cooperation and demonstrate that the Nash equilibria depend not only on the preference alignment but also on whether the players need cooperation to achieve a better outcome and whether they are willing to cooperate.We illustrate the theoretical results by solving a mechanism design problem for a drone delivery scenario., Comment: 14 page, 6 figure, under development

This work proposes a mathematical framework to increase the robustness to rare events of digital twins modelled with graphical models. We incorporate probabilistic model-checking and linear programming into a dynamic Bayesian network to enable the construction of risk-averse digital twins. By modeling with a random variable the probability of the asset to transition from one state to another, we define a parametric Markov decision process. By solving this Markov decision process, we compute a policy that defines state-dependent optimal actions to take. To account for rare events connected to failures we leverage risk measures associated with the distribution of the random variables describing the transition probabilities. We refine the optimal policy at every time step resulting in a better trade off between operational costs and performances. We showcase the capabilities of the proposed framework with a structural digital twin of an unmanned aerial vehicle and its adaptive mission replanning.

Deception is helpful for agents masking their intentions from an observer. We consider a team of agents deceiving their supervisor. The supervisor defines nominal behavior for the agents via reference policies, but the agents share an alternate task that they can only achieve by deviating from these references. As such, the agents use deceptive policies to complete the task while ensuring that their behaviors remain plausible to the supervisor. We propose a setting with centralized deceptive policy synthesis and decentralized execution. We model each agent with a Markov decision process and constrain the agents' deceptive policies so that, with high probability, at least one agent achieves the task. We then provide an algorithm to synthesize deceptive policies that ensure the deviations of all agents are small by minimizing the worst Kullback-Leibler divergence between any agent's deceptive and reference policies. Thanks to decentralization, this algorithm scales linearly with the number of agents and also facilitates the efficient synthesis of reference policies. We then explore a more general version of the deceptive policy synthesis problem. In particular, we consider a supervisor who selects a subset of agents to eliminate based on the agents' behaviors. We give algorithms to synthesize deceptive policies so that, after the supervisor eliminates some agents, the remaining agents complete the task with high probability. We demonstrate the developed methods in a package delivery example., Comment: 20 pages, 5 figures Accepted at the Conference on Game Theory and AI for Security (GameSec 2024)

Zero-sum games arise in a wide variety of problems, including robust optimization and adversarial learning. However, algorithms deployed for finding a local Nash equilibrium in these games often converge to non-Nash stationary points. This highlights a key challenge: for any algorithm, the stability properties of its underlying dynamical system can cause non-Nash points to be potential attractors. To overcome this challenge, algorithms must account for subtleties involving the curvatures of players' costs. To this end, we leverage dynamical system theory and develop a second-order algorithm for finding a local Nash equilibrium in the smooth, possibly nonconvex-nonconcave, zero-sum game setting. First, we prove that this novel method guarantees convergence to only local Nash equilibria with a local linear convergence rate. We then interpret a version of this method as a modified Gauss-Newton algorithm with local superlinear convergence to the neighborhood of a point that satisfies first-order local Nash equilibrium conditions. In comparison, current related state-of-the-art methods do not offer convergence rate guarantees. Furthermore, we show that this approach naturally generalizes to settings with convex and potentially coupled constraints while retaining earlier guarantees of convergence to only local (generalized) Nash equilibria.

Developing autonomous agents that can strategize and cooperate with humans under information asymmetry is challenging without effective communication in natural language. We introduce a shared-control game, where two players collectively control a token in alternating turns to achieve a common objective under incomplete information. We formulate a policy synthesis problem for an autonomous agent in this game with a human as the other player. To solve this problem, we propose a communication-based approach comprising a language module and a planning module. The language module translates natural language messages into and from a finite set of flags, a compact representation defined to capture player intents. The planning module leverages these flags to compute a policy using an asymmetric information-set Monte Carlo tree search with flag exchange algorithm we present. We evaluate the effectiveness of this approach in a testbed based on Gnomes at Night, a search-and-find maze board game. Results of human subject experiments show that communication narrows the information gap between players and enhances human-agent cooperation efficiency with fewer turns., Comment: with appendix

Autonomous systems often encounter environments and scenarios beyond the scope of their training data, which underscores a critical challenge: the need to generalize and adapt to unseen scenarios in real time. This challenge necessitates new mathematical and algorithmic tools that enable adaptation and zero-shot transfer. To this end, we leverage the theory of function encoders, which enables zero-shot transfer by combining the flexibility of neural networks with the mathematical principles of Hilbert spaces. Using this theory, we first present a method for learning a space of dynamics spanned by a set of neural ODE basis functions. After training, the proposed approach can rapidly identify dynamics in the learned space using an efficient inner product calculation. Critically, this calculation requires no gradient calculations or retraining during the online phase. This method enables zero-shot transfer for autonomous systems at runtime and opens the door for a new class of adaptable control algorithms. We demonstrate state-of-the-art system modeling accuracy for two MuJoCo robot environments and show that the learned models can be used for more efficient MPC control of a quadrotor.

We study a pair of budget- and performance-constrained weak submodular maximization problems. For computational efficiency, we explore the use of stochastic greedy algorithms which limit the search space via random sampling instead of the standard greedy procedure which explores the entire feasible search space. We propose a pair of stochastic greedy algorithms, namely, Modified Randomized Greedy (MRG) and Dual Randomized Greedy (DRG) to approximately solve the budget- and performance-constrained problems, respectively. For both algorithms, we derive approximation guarantees that hold with high probability. We then examine the use of DRG in robust optimization problems wherein the objective is to maximize the worst-case of a number of weak submodular objectives and propose the Randomized Weak Submodular Saturation Algorithm (Random-WSSA). We further derive a high-probability guarantee for when Random-WSSA successfully constructs a robust solution. Finally, we showcase the effectiveness of these algorithms in a variety of relevant uses within the context of Earth-observing LEO constellations which estimate atmospheric weather conditions and provide Earth coverage., Comment: 36 pages, 5 figures. A preliminary version of this article was presented at the 2023 American Control Conference (ACC). This version was submitted to Automatica

Data-poisoning attacks can disrupt the efficient operations of transportation systems by misdirecting traffic flows via falsified data. One challenge in countering these attacks is to reduce the uncertainties on the types of attacks, such as the distribution of their targets and intensities. We introduce a resource allocation method in transportation networks to detect and distinguish different types of attacks and facilitate efficient traffic routing. The idea is to first cluster different types of attacks based on the corresponding optimal routing strategies, then allocate sensing resources to a subset of network links to distinguish attacks from different clusters via lexicographical mixed-integer programming. We illustrate the application of the proposed method using the Anaheim network, a benchmark model in traffic routing that contains more than 400 nodes and 900 links.

Simultaneous localization and mapping is essential for position tracking and scene understanding. 3D Gaussian-based map representations enable photorealistic reconstruction and real-time rendering of scenes using multiple posed cameras. We show for the first time that using 3D Gaussians for map representation with unposed camera images and inertial measurements can enable accurate SLAM. Our method, MM3DGS, addresses the limitations of prior neural radiance field-based representations by enabling faster rendering, scale awareness, and improved trajectory tracking. Our framework enables keyframe-based mapping and tracking utilizing loss functions that incorporate relative pose transformations from pre-integrated inertial measurements, depth estimates, and measures of photometric rendering quality. We also release a multi-modal dataset, UT-MM, collected from a mobile robot equipped with a camera and an inertial measurement unit. Experimental evaluation on several scenes from the dataset shows that MM3DGS achieves 3x improvement in tracking and 5% improvement in photometric rendering quality compared to the current 3DGS SLAM state-of-the-art, while allowing real-time rendering of a high-resolution dense 3D map. Project Webpage: https://vita-group.github.io/MM3DGS-SLAM, Comment: Project Webpage: https://vita-group.github.io/MM3DGS-SLAM

We present an active learning algorithm for learning dynamics that leverages side information by explicitly incorporating prior domain knowledge into the sampling process. Our proposed algorithm guides the exploration toward regions that demonstrate high empirical discrepancy between the observed data and an imperfect prior model of the dynamics derived from side information. Through numerical experiments, we demonstrate that this strategy explores regions of high discrepancy and accelerates learning while simultaneously reducing model uncertainty. We rigorously prove that our active learning algorithm yields a consistent estimate of the underlying dynamics by providing an explicit rate of convergence for the maximum predictive variance. We demonstrate the efficacy of our approach on an under-actuated pendulum system and on the half-cheetah MuJoCo environment.

Based on our workshop activities, we outlined three ways in which research can support community needs: (1) Mapping the ecosystem of both the players and ecosystem and harm landscapes, (2) Counter-Programming, which entails using the same surveillance tools that communities are subjected to observe the entities doing the surveilling, effectively protecting people from surveillance, and conducting ethical data collection to measure the impact of these technologies, and (3) Engaging in positive visions and tools for empowerment so that technology can bring good instead of harm. In order to effectively collaborate on the aforementioned directions, we outlined seven important mechanisms for effective collaboration: (1) Never expect free labor of community members, (2) Ensure goals are aligned between all collaborators, (3) Elevate community members to leadership positions, (4) Understand no group is a monolith, (5) Establish a common language, (6) Discuss organization roles and goals of the project transparently from the start, and (7) Enable a recourse for harm. We recommend that anyone engaging in community-based research (1) starts with community-defined solutions, (2) provides alternatives to digital services/information collecting mechanisms, (3) prohibits harmful automated systems, (4) transparently states any systems impact, (5) minimizes and protects data, (6) proactively demonstrates a system is safe and beneficial prior to deployment, and (7) provides resources directly to community partners. Throughout the recommendation section of the report, we also provide specific recommendations for funding agencies, academic institutions, and individual researchers.

We address the problem of sparse selection of visual features for localizing a team of robots navigating an unknown environment, where robots can exchange relative position measurements with neighbors. We select a set of the most informative features by anticipating their importance in robots localization by simulating trajectories of robots over a prediction horizon. Through theoretical proofs, we establish a crucial connection between graph Laplacian and the importance of features. We show that strong network connectivity translates to uniformity in feature importance, which enables uniform random sampling of features and reduces the overall computational complexity. We leverage a scalable randomized algorithm for sparse sums of positive semidefinite matrices to efficiently select the set of the most informative features and significantly improve the probabilistic performance bounds. Finally, we support our findings with extensive simulations.

Social news websites, such as Reddit, have evolved into prominent platforms for sharing and discussing news. A key issue on social news websites sites is the formation of echo chambers, which often lead to the spread of highly biased or uncredible news. We develop a method to identify communities within a social news website that are prone to uncredible or highly biased news. We employ a user embedding pipeline that detects user communities based on their stances towards posts and news sources. We then project each community onto a credibility-bias space and analyze the distributional characteristics of each projected community to identify those that have a high risk of adopting beliefs with low credibility or high bias. This approach also enables the prediction of individual users' susceptibility to low credibility content, based on their community affiliation. Our experiments show that latent space clusters effectively indicate the credibility and bias levels of their users, with significant differences observed across clusters -- a $34\%$ difference in the users' susceptibility to low-credibility content and a $8.3\%$ difference in the users' susceptibility to high political bias., Comment: 11 pages, 2 figures, Under review in ICWSM 2024

Coordination in multiplayer games enables players to avoid the lose-lose outcome that often arises at Nash equilibria. However, designing a coordination mechanism typically requires the consideration of the joint actions of all players, which becomes intractable in large-scale games. We develop a novel coordination mechanism, termed reduced rank correlated equilibria, which reduces the number of joint actions to be considered and thereby mitigates computational complexity. The idea is to approximate the set of all joint actions with the actions used in a set of pre-computed Nash equilibria via a convex hull operation. In a game with n players and each player having m actions, the proposed mechanism reduces the number of joint actions considered from O(m^n) to O(mn). We demonstrate the application of the proposed mechanism to an air traffic queue management problem. Compared with the correlated equilibrium-a popular benchmark coordination mechanism-the proposed approach is capable of solving a problem involving four thousand times more joint actions while yielding similar or better performance in terms of a fairness indicator and showing a maximum optimality gap of 0.066% in terms of the average delay cost. In the meantime, it yields a solution that shows up to 99.5% improvement in a fairness indicator and up to 50.4% reduction in average delay cost compared to the Nash solution, which does not involve coordination.

Foundation models have recently expanded into robotics after excelling in computer vision and natural language processing. The models are accessible in two ways: open-source or paid, closed-source options. Users with access to both face a problem when deciding between effective yet costly closed-source models and free but less powerful open-source alternatives. We call it the model selection problem. Existing supervised-learning methods are impractical due to the high cost of collecting extensive training data from closed-source models. Hence, we focus on the online learning setting where algorithms learn while collecting data, eliminating the need for large pre-collected datasets. We thus formulate a user-centric online model selection problem and propose a novel solution that combines an open-source encoder to output context and an online learning algorithm that processes this context. The encoder distills vast data distributions into low-dimensional features, i.e., the context, without additional training. The online learning algorithm aims to maximize a composite reward that includes model performance, execution time, and costs based on the context extracted from the data. It results in an improved trade-off between selecting open-source and closed-source models compared to non-contextual methods, as validated by our theoretical analysis. Experiments across language-based robotic tasks such as Waymo Open Dataset, ALFRED, and Open X-Embodiment demonstrate real-world applications of the solution. The results show that the solution significantly improves the task success rate by up to 14%.

When multiple agents interact in a common environment, each agent's actions impact others' future decisions, and noncooperative dynamic games naturally capture this coupling. In interactive motion planning, however, agents typically do not have access to a complete model of the game, e.g., due to unknown objectives of other players. Therefore, we consider the inverse game problem, in which some properties of the game are unknown a priori and must be inferred from observations. Existing maximum likelihood estimation (MLE) approaches to solve inverse games provide only point estimates of unknown parameters without quantifying uncertainty, and perform poorly when many parameter values explain the observed behavior. To address these limitations, we take a Bayesian perspective and construct posterior distributions of game parameters. To render inference tractable, we employ a variational autoencoder (VAE) with an embedded differentiable game solver. This structured VAE can be trained from an unlabeled dataset of observed interactions, naturally handles continuous, multi-modal distributions, and supports efficient sampling from the inferred posteriors without computing game solutions at runtime. Extensive evaluations in simulated driving scenarios demonstrate that the proposed approach successfully learns the prior and posterior game parameter distributions, provides more accurate objective estimates than MLE baselines, and facilitates safer and more efficient game-theoretic motion planning.

We present LARL-RM (Large language model-generated Automaton for Reinforcement Learning with Reward Machine) algorithm in order to encode high-level knowledge into reinforcement learning using automaton to expedite the reinforcement learning. Our method uses Large Language Models (LLM) to obtain high-level domain-specific knowledge using prompt engineering instead of providing the reinforcement learning algorithm directly with the high-level knowledge which requires an expert to encode the automaton. We use chain-of-thought and few-shot methods for prompt engineering and demonstrate that our method works using these approaches. Additionally, LARL-RM allows for fully closed-loop reinforcement learning without the need for an expert to guide and supervise the learning since LARL-RM can use the LLM directly to generate the required high-level knowledge for the task at hand. We also show the theoretical guarantee of our algorithm to converge to an optimal policy. We demonstrate that LARL-RM speeds up the convergence by 30% by implementing our method in two case studies.

In the era of social media platforms, identifying the credibility of online content is crucial to combat misinformation. We present the CREDiBERT (CREDibility assessment using Bi-directional Encoder Representations from Transformers), a source credibility assessment model fine-tuned for Reddit submissions focusing on political discourse as the main contribution. We adopt a semi-supervised training approach for CREDiBERT, leveraging Reddit's community-based structure. By encoding submission content using CREDiBERT and integrating it into a Siamese neural network, we significantly improve the binary classification of submission credibility, achieving a 9% increase in F1 score compared to existing methods. Additionally, we introduce a new version of the post-to-post network in Reddit that efficiently encodes user interactions to enhance the binary classification task by nearly 8% in F1 score. Finally, we employ CREDiBERT to evaluate the susceptibility of subreddits with respect to different topics., Comment: 12 pages; 3 figures

Although reinforcement learning (RL) can solve many challenging sequential decision making problems, achieving zero-shot transfer across related tasks remains a challenge. The difficulty lies in finding a good representation for the current task so that the agent understands how it relates to previously seen tasks. To achieve zero-shot transfer, we introduce the function encoder, a representation learning algorithm which represents a function as a weighted combination of learned, non-linear basis functions. By using a function encoder to represent the reward function or the transition function, the agent has information on how the current task relates to previously seen tasks via a coherent vector representation. Thus, the agent is able to achieve transfer between related tasks at run time with no additional training. We demonstrate state-of-the-art data efficiency, asymptotic performance, and training stability in three RL fields by augmenting basic RL algorithms with a function encoder task representation.

Urban air mobility (UAM), a transformative concept for the transport of passengers and cargo, faces several integration challenges in complex urban environments. Community acceptance of aircraft noise is among the most noticeable of these challenges when launching or scaling up a UAM system. Properly managing community noise is fundamental to establishing a UAM system that is environmentally and socially sustainable. In this work, we develop a holistic and equitable approach to manage UAM air traffic and its community noise impact in urban environments. The proposed approach is a hybrid approach that considers a mix of different noise mitigation strategies, including limiting the number of operations, cruising at higher altitudes, and ambient noise masking. We tackle the problem through the lens of network system control and formulate a multi-objective optimization model for managing traffic flow in a multi-layer UAM network while concurrently pursuing demand fulfillment, noise control, and energy saving. Further, we use a social welfare function in the optimization model as the basis for the efficiency-fairness trade-off in both demand fulfillment and noise control. We apply the proposed approach to a comprehensive case study in the city of Austin and perform design trade-offs through both visual and quantitative analyses., Comment: 30 pages, 15 figures

Securing dynamic networks against adversarial actions is challenging because of the need to anticipate and counter strategic disruptions by adversarial entities within complex network structures. Traditional game-theoretic models, while insightful, often fail to model the unpredictability and constraints of real-world threat assessment scenarios. We refine sabotage games to reflect the realistic limitations of the saboteur and the network operator. By transforming sabotage games into reachability problems, our approach allows applying existing computational solutions to model realistic restrictions on attackers and defenders within the game. Modifying sabotage games into dynamic network security problems successfully captures the nuanced interplay of strategy and uncertainty in dynamic network security. Theoretically, we extend sabotage games to model network security contexts and thoroughly explore if the additional restrictions raise their computational complexity, often the bottleneck of game theory in practical contexts. Practically, this research sets the stage for actionable insights for developing robust defense mechanisms by understanding what risks to mitigate in dynamically changing networks under threat.

We propose and demonstrate a compositional framework for training and verifying reinforcement learning (RL) systems within a multifidelity sim-to-real pipeline, in order to deploy reliable and adaptable RL policies on physical hardware. By decomposing complex robotic tasks into component subtasks and defining mathematical interfaces between them, the framework allows for the independent training and testing of the corresponding subtask policies, while simultaneously providing guarantees on the overall behavior that results from their composition. By verifying the performance of these subtask policies using a multifidelity simulation pipeline, the framework not only allows for efficient RL training, but also for a refinement of the subtasks and their interfaces in response to challenges arising from discrepancies between simulation and reality. In an experimental case study we apply the framework to train and deploy a compositional RL system that successfully pilots a Warthog unmanned ground robot.

The growing reliance on social media for news consumption necessitates effective countermeasures to mitigate the rapid spread of misinformation. Prebunking, a proactive method that arms users with accurate information before they come across false content, has garnered support from journalism and psychology experts. We formalize the problem of optimal prebunking as optimizing the timing of delivering accurate information, ensuring users encounter it before receiving misinformation while minimizing the disruption to user experience. Utilizing a susceptible-infected epidemiological process to model the propagation of misinformation, we frame optimal prebunking as a policy synthesis problem with safety constraints. We then propose a policy that approximates the optimal solution to a relaxed problem. The experiments show that this policy cuts the user experience cost of repeated information delivery in half, compared to delivering accurate information immediately after identifying a misinformation propagation., Comment: 10 pages, 3 figures, Submitted to PERCOM 2024

Formal methods refer to rigorous, mathematical approaches to system development and have played a key role in establishing the correctness of safety-critical systems. The main building blocks of formal methods are models and specifications, which are analogous to behaviors and requirements in system design and give us the means to verify and synthesize system behaviors with formal guarantees. This monograph provides a survey of the current state of the art on applications of formal methods in the autonomous systems domain. We consider correct-by-construction synthesis under various formulations, including closed systems, reactive, and probabilistic settings. Beyond synthesizing systems in known environments, we address the concept of uncertainty and bound the behavior of systems that employ learning using formal methods. Further, we examine the synthesis of systems with monitoring, a mitigation technique for ensuring that once a system deviates from expected behavior, it knows a way of returning to normalcy. We also show how to overcome some limitations of formal methods themselves with learning. We conclude with future directions for formal methods in reinforcement learning, uncertainty, privacy, explainability of formal methods, and regulation and certification.

Although pre-trained language models encode generic knowledge beneficial for planning and control, they may fail to generate appropriate control policies for domain-specific tasks. Existing fine-tuning methods use human feedback to address this limitation, however, sourcing human feedback is labor intensive and costly. We present a fully automated approach to fine-tune pre-trained language models for applications in autonomous systems, bridging the gap between generic knowledge and domain-specific requirements while reducing cost. The method synthesizes automaton-based controllers from pre-trained models guided by natural language task descriptions. These controllers are verifiable against independently provided specifications within a world model, which can be abstract or obtained from a high-fidelity simulator. Controllers with high compliance with the desired specifications receive higher ranks, guiding the iterative fine-tuning process. We provide quantitative evidences, primarily in autonomous driving, to demonstrate the method's effectiveness across multiple tasks. The results indicate an improvement in percentage of specifications satisfied by the controller from 60% to 90%.

Algorithmic robustness refers to the sustained performance of a computational system in the face of change in the nature of the environment in which that system operates or in the task that the system is meant to perform. Below, we motivate the importance of algorithmic robustness, present a conceptual framework, and highlight the relevant areas of research for which algorithmic robustness is relevant. Why robustness? Robustness is an important enabler of other goals that are frequently cited in the context of public policy decisions about computational systems, including trustworthiness, accountability, fairness, and safety. Despite this dependence, it tends to be under-recognized compared to these other concepts. This is unfortunate, because robustness is often more immediately achievable than these other ultimate goals, which can be more subjective and exacting. Thus, we highlight robustness as an important goal for researchers, engineers, regulators, and policymakers when considering the design, implementation, and deployment of computational systems. We urge researchers and practitioners to elevate the attention paid to robustness when designing and evaluating computational systems. For many key systems, the immediate question after any demonstration of high performance should be: "How robust is that performance to realistic changes in the task or environment?" Greater robustness will set the stage for systems that are more trustworthy, accountable, fair, and safe. Toward that end, this document provides a brief roadmap to some of the concepts and existing research around the idea of algorithmic robustness.

When interacting with other non-competitive decision-making agents, it is critical for an autonomous agent to have inferable behavior: Their actions must convey their intention and strategy. For example, an autonomous car's strategy must be inferable by the pedestrians interacting with the car. We model the inferability problem using a repeated bimatrix Stackelberg game with observations where a leader and a follower repeatedly interact. During the interactions, the leader uses a fixed, potentially mixed strategy. The follower, on the other hand, does not know the leader's strategy and dynamically reacts based on observations that are the leader's previous actions. In the setting with observations, the leader may suffer from an inferability loss, i.e., the performance compared to the setting where the follower has perfect information of the leader's strategy. We show that the inferability loss is upper-bounded by a function of the number of interactions and the stochasticity level of the leader's strategy, encouraging the use of inferable strategies with lower stochasticity levels. As a converse result, we also provide a game where the required number of interactions is lower bounded by a function of the desired inferability loss.

In the vision of urban air mobility, air transport systems serve the demands of urban communities by routing flight traffic in networks formed by vertiports and flight corridors. We develop a routing algorithm to ensure that the air traffic flow fairly serves the demand of multiple communities subject to stochastic network capacity constraints. This algorithm guarantees that the flight traffic volume allocated to different communities satisfies the \emph{alpha-fairness conditions}, a commonly used family of fairness conditions in resource allocation. It further ensures robust satisfaction of stochastic network capacity constraints by bounding the coherent risk measures of capacity violation. We prove that implementing the proposed algorithm is equivalent to solving a convex optimization problem. We demonstrate the proposed algorithm using a case study based on the city of Austin. Compared with one that maximizes the total served demands, the proposed algorithm promotes even distributions of served demands for different communities.

The increasing abundance of video data enables users to search for events of interest, e.g., emergency incidents. Meanwhile, it raises new concerns, such as the need for preserving privacy. Existing approaches to video search require either manual inspection or a deep learning model with massive training. We develop a method that uses recent advances in vision and language models, as well as formal methods, to search for events of interest in video clips automatically and efficiently. The method consists of an algorithm to map text-based event descriptions into linear temporal logic over finite traces (LTL$_f$) and an algorithm to construct an automaton encoding the video information. Then, the method formally verifies the automaton representing the video against the LTL$_f$ specifications and adds the pertinent video clips to the search result if the automaton satisfies the specifications. We provide qualitative and quantitative analysis to demonstrate the video-searching capability of the proposed method. It achieves over 90 percent precision in searching over privacy-sensitive videos and a state-of-the-art autonomous driving dataset., Comment: 12 pages, 18 figures

In cooperative multi-agent reinforcement learning (Co-MARL), a team of agents must jointly optimize the team's long-term rewards to learn a designated task. Optimizing rewards as a team often requires inter-agent communication and data sharing, leading to potential privacy implications. We assume privacy considerations prohibit the agents from sharing their environment interaction data. Accordingly, we propose Privacy-Engineered Value Decomposition Networks (PE-VDN), a Co-MARL algorithm that models multi-agent coordination while provably safeguarding the confidentiality of the agents' environment interaction data. We integrate three privacy-engineering techniques to redesign the data flows of the VDN algorithm, an existing Co-MARL algorithm that consolidates the agents' environment interaction data to train a central controller that models multi-agent coordination, and develop PE-VDN. In the first technique, we design a distributed computation scheme that eliminates Vanilla VDN's dependency on sharing environment interaction data. Then, we utilize a privacy-preserving multi-party computation protocol to guarantee that the data flows of the distributed computation scheme do not pose new privacy risks. Finally, we enforce differential privacy to preempt inference threats against the agents' training data, past environment interactions, when they take actions based on their neural network predictions. We implement PE-VDN in StarCraft Multi-Agent Competition (SMAC) and show that it achieves 80% of Vanilla VDN's win rate while maintaining differential privacy levels that provide meaningful privacy guarantees. The results demonstrate that PE-VDN can safeguard the confidentiality of agents' environment interaction data without sacrificing multi-agent coordination., Comment: Paper accepted at 62nd IEEE Conference on Decision and Control

We propose a framework for verifiable and compositional reinforcement learning (RL) in which a collection of RL subsystems, each of which learns to accomplish a separate subtask, are composed to achieve an overall task. The framework consists of a high-level model, represented as a parametric Markov decision process, which is used to plan and analyze compositions of subsystems, and of the collection of low-level subsystems themselves. The subsystems are implemented as deep RL agents operating under partial observability. By defining interfaces between the subsystems, the framework enables automatic decompositions of task specifications, e.g., reach a target set of states with a probability of at least 0.95, into individual subtask specifications, i.e. achieve the subsystem's exit conditions with at least some minimum probability, given that its entry conditions are met. This in turn allows for the independent training and testing of the subsystems. We present theoretical results guaranteeing that if each subsystem learns a policy satisfying its subtask specification, then their composition is guaranteed to satisfy the overall task specification. Conversely, if the subtask specifications cannot all be satisfied by the learned policies, we present a method, formulated as the problem of finding an optimal set of parameters in the high-level model, to automatically update the subtask specifications to account for the observed shortcomings. The result is an iterative procedure for defining subtask specifications, and for training the subsystems to meet them. Experimental results demonstrate the presented framework's novel capabilities in environments with both full and partial observability, discrete and continuous state and action spaces, as well as deterministic and stochastic dynamics., Comment: arXiv admin note: substantial text overlap with arXiv:2106.05864

We consider a setting where a supervisor delegates an agent to perform a certain control task, while the agent is incentivized to deviate from the given policy to achieve its own goal. In this work, we synthesize the optimal deceptive policies for an agent who attempts to hide its deviations from the supervisor's policy. We study the deception problem in the continuous-state discrete-time stochastic dynamics setting and, using motivations from hypothesis testing theory, formulate a Kullback-Leibler control problem for the synthesis of deceptive policies. This problem can be solved using backward dynamic programming in principle, which suffers from the curse of dimensionality. However, under the assumption of deterministic state dynamics, we show that the optimal deceptive actions can be generated using path integral control. This allows the agent to numerically compute the deceptive actions via Monte Carlo simulations. Since Monte Carlo simulations can be efficiently parallelized, our approach allows the agent to generate deceptive control actions online. We show that the proposed simulation-driven control approach asymptotically converges to the optimal control distribution., Comment: 8 pages, 3 figures, CDC 2023

Game-theoretic inverse learning is the problem of inferring the players' objectives from their actions. We formulate an inverse learning problem in a Stackelberg game between a leader and a follower, where each player's action is the trajectory of a dynamical system. We propose an active inverse learning method for the leader to infer which hypothesis among a finite set of candidates describes the follower's objective function. Instead of using passively observed trajectories like existing methods, the proposed method actively maximizes the differences in the follower's trajectories under different hypotheses to accelerate the leader's inference. We demonstrate the proposed method in a receding-horizon repeated trajectory game. Compared with uniformly random inputs, the leader inputs provided by the proposed method accelerate the convergence of the probability of different hypotheses conditioned on the follower's trajectory by orders of magnitude.

Recently developed pretrained models can encode rich world knowledge expressed in multiple modalities, such as text and images. However, the outputs of these models cannot be integrated into algorithms to solve sequential decision-making tasks. We develop an algorithm that utilizes the knowledge from pretrained models to construct and verify controllers for sequential decision-making tasks, and to ground these controllers to task environments through visual observations with formal guarantees. In particular, the algorithm queries a pretrained model with a user-provided, text-based task description and uses the model's output to construct an automaton-based controller that encodes the model's task-relevant knowledge. It allows formal verification of whether the knowledge encoded in the controller is consistent with other independently available knowledge, which may include abstract information on the environment or user-provided specifications. Next, the algorithm leverages the vision and language capabilities of pretrained models to link the observations from the task environment to the text-based control logic from the controller (e.g., actions and conditions that trigger the actions). We propose a mechanism to provide probabilistic guarantees on whether the controller satisfies the user-provided specifications under perceptual uncertainties. We demonstrate the algorithm's ability to construct, verify, and ground automaton-based controllers through a suite of real-world tasks, including daily life and robot manipulation tasks., Comment: Accepted as full paper in AAMAS 2024

We study a class of reinforcement learning (RL) tasks where the objective of the agent is to accomplish temporally extended goals. In this setting, a common approach is to represent the tasks as deterministic finite automata (DFA) and integrate them into the state-space for RL algorithms. However, while these machines model the reward function, they often overlook the causal knowledge about the environment. To address this limitation, we propose the Temporal-Logic-based Causal Diagram (TL-CD) in RL, which captures the temporal causal relationships between different properties of the environment. We exploit the TL-CD to devise an RL algorithm in which an agent requires significantly less exploration of the environment. To this end, based on a TL-CD and a task DFA, we identify configurations where the agent can determine the expected rewards early during an exploration. Through a series of case studies, we demonstrate the benefits of using TL-CDs, particularly the faster convergence of the algorithm to an optimal policy due to reduced exploration of the environment.

We present a framework and algorithms to learn controlled dynamics models using neural stochastic differential equations (SDEs) -- SDEs whose drift and diffusion terms are both parametrized by neural networks. We construct the drift term to leverage a priori physics knowledge as inductive bias, and we design the diffusion term to represent a distance-aware estimate of the uncertainty in the learned model's predictions -- it matches the system's underlying stochasticity when evaluated on states near those from the training dataset, and it predicts highly stochastic dynamics when evaluated on states beyond the training regime. The proposed neural SDEs can be evaluated quickly enough for use in model predictive control algorithms, or they can be used as simulators for model-based reinforcement learning. Furthermore, they make accurate predictions over long time horizons, even when trained on small datasets that cover limited regions of the state space. We demonstrate these capabilities through experiments on simulated robotic systems, as well as by using them to model and control a hexacopter's flight dynamics: A neural SDE trained using only three minutes of manually collected flight data results in a model-based control policy that accurately tracks aggressive trajectories that push the hexacopter's velocity and Euler angles to nearly double the maximum values observed in the training dataset., Comment: Final submission to CoRL 2023

Near the limits of adhesion, the forces generated by a tire are nonlinear and intricately coupled. Efficient and accurate modelling in this region could improve safety, especially in emergency situations where high forces are required. To this end, we propose a novel family of tire force models based on neural ordinary differential equations and a neural-ExpTanh parameterization. These models are designed to satisfy physically insightful assumptions while also having sufficient fidelity to capture higher-order effects directly from vehicle state measurements. They are used as drop-in replacements for an analytical brush tire model in an existing nonlinear model predictive control framework. Experiments with a customized Toyota Supra show that scarce amounts of driving data -- less than three minutes -- is sufficient to achieve high-performance autonomous drifting on various trajectories with speeds up to 45mph. Comparisons with the benchmark model show a $4 \times$ improvement in tracking performance, smoother control inputs, and faster and more consistent computation time., Comment: Final Submission at ICRA 2023

Urban air mobility (UAM), as envisioned by aviation professionals, will transport passengers and cargo at low altitudes within urban and suburban areas. To operate in urban environments, precise air traffic management, in particular the management of traffic overflows due to physical and operational disruptions will be critical to ensuring system safety and efficiency. To this end, we propose UAM network design with reserve capacity, i.e., a design where alternative landing options and flight corridors are explicitly considered as a means of improving contingency management. Similar redundancy considerations are incorporated in the design of many critical infrastructures, yet remain unexploited in the air transportation literature. In our methodology, we first model how disruptions to a given UAM network might impact on the nominal traffic flow and how this flow might be re-accommodated on an extended network with reserve capacity. Then, through an optimization problem, we select the locations and capacities for the backup vertiports with the maximal expected throughput of the extended network over all possible disruption scenarios, while the throughput is the maximal amount of flights that the network can accommodate per unit of time. We show that we can obtain the solution for the corresponding bi-level and bi-linear optimization problem by solving a mixed-integer linear program. We demonstrate our methodology in the case study using networks from Milwaukee, Atlanta, and Dallas--Fort Worth metropolitan areas and show how the throughput and flexibility of the UAM networks with reserve capacity can outcompete those without., Comment: 44 pages, 10 figures

We investigate multi-agent reinforcement learning for stochastic games with complex tasks, where the reward functions are non-Markovian. We utilize reward machines to incorporate high-level knowledge of complex tasks. We develop an algorithm called Q-learning with reward machines for stochastic games (QRM-SG), to learn the best-response strategy at Nash equilibrium for each agent. In QRM-SG, we define the Q-function at a Nash equilibrium in augmented state space. The augmented state space integrates the state of the stochastic game and the state of reward machines. Each agent learns the Q-functions of all agents in the system. We prove that Q-functions learned in QRM-SG converge to the Q-functions at a Nash equilibrium if the stage game at each time step during learning has a global optimum point or a saddle point, and the agents update Q-functions based on the best-response strategy at this point. We use the Lemke-Howson method to derive the best-response strategy given current Q-functions. The three case studies show that QRM-SG can learn the best-response strategies effectively. QRM-SG learns the best-response strategies after around 7500 episodes in Case Study I, 1000 episodes in Case Study II, and 1500 episodes in Case Study III, while baseline methods such as Nash Q-learning and MADDPG fail to converge to the Nash equilibrium in all three case studies.

Self-paced reinforcement learning (RL) aims to improve the data efficiency of learning by automatically creating sequences, namely curricula, of probability distributions over contexts. However, existing techniques for self-paced RL fail in long-horizon planning tasks that involve temporally extended behaviors. We hypothesize that taking advantage of prior knowledge about the underlying task structure can improve the effectiveness of self-paced RL. We develop a self-paced RL algorithm guided by reward machines, i.e., a type of finite-state machine that encodes the underlying task structure. The algorithm integrates reward machines in 1) the update of the policy and value functions obtained by any RL algorithm of choice, and 2) the update of the automated curriculum that generates context distributions. Our empirical results evidence that the proposed algorithm achieves optimal behavior reliably even in cases in which existing baselines cannot make any meaningful progress. It also decreases the curriculum length and reduces the variance in the curriculum generation process by up to one-fourth and four orders of magnitude, respectively., Comment: 9 pages, 11 figures. Accepted for UAI 2023

Efficient compression of correlated data is essential to minimize communication overload in multi-sensor networks. In such networks, each sensor independently compresses the data and transmits them to a central node due to limited communication bandwidth. A decoder at the central node decompresses and passes the data to a pre-trained machine learning-based task to generate the final output. Thus, it is important to compress the features that are relevant to the task. Additionally, the final performance depends heavily on the total available bandwidth. In practice, it is common to encounter varying availability in bandwidth, and higher bandwidth results in better performance of the task. We design a novel distributed compression framework composed of independent encoders and a joint decoder, which we call neural distributed principal component analysis (NDPCA). NDPCA flexibly compresses data from multiple sources to any available bandwidth with a single model, reducing computing and storage overhead. NDPCA achieves this by learning low-rank task representations and efficiently distributing bandwidth among sensors, thus providing a graceful trade-off between performance and bandwidth. Experiments show that NDPCA improves the success rate of multi-view robotic arm manipulation by 9% and the accuracy of object detection tasks on satellite imagery by 14% compared to an autoencoder with uniform bandwidth allocation.

Urban air mobility (UAM) is an emerging concept in short-range aviation transportation, where the aircraft will take off, land, and charge their batteries at a set of vertistops, and travel only through a set of flight corridors connecting these vertistops. We study the problem of routing an electric aircraft from its origin vertistop to its destination vertistop with the minimal expected total travel time. We first introduce a UAM network model that accounts for the limited battery capacity of aircraft, stochastic travel times of flight corridors, stochastic queueing delays, and a limited number of battery-charging stations at vertistops. Based on this model, we provide a sufficient condition for the existence of a routing strategy that avoids battery exhaustion. Furthermore, we show how to compute such a strategy by computing the optimal policy in a Markov decision process, a mathematical framework for decision-making in a stochastic dynamic environment. We illustrate our results using a case study with 29 vertistops and 137 flight corridors., Comment: 8 pages, 3 figures

We provide a novel method for sensitivity analysis of parametric robust Markov chains. These models incorporate parameters and sets of probability distributions to alleviate the often unrealistic assumption that precise probabilities are available. We measure sensitivity in terms of partial derivatives with respect to the uncertain transition probabilities regarding measures such as the expected reward. As our main contribution, we present an efficient method to compute these partial derivatives. To scale our approach to models with thousands of parameters, we present an extension of this method that selects the subset of $k$ parameters with the highest partial derivative. Our methods are based on linear programming and differentiating these programs around a given value for the parameters. The experiments show the applicability of our approach on models with over a million states and thousands of parameters. Moreover, we embed the results within an iterative learning scheme that profits from having access to a dedicated sensitivity analysis., Comment: To be presented at CAV 2023

Markov games model interactions among multiple players in a stochastic, dynamic environment. Each player in a Markov game maximizes its expected total discounted reward, which depends upon the policies of the other players. We formulate a class of Markov games, termed affine Markov games, where an affine reward function couples the players' actions. We introduce a novel solution concept, the soft-Bellman equilibrium, where each player is boundedly rational and chooses a soft-Bellman policy rather than a purely rational policy as in the well-known Nash equilibrium concept. We provide conditions for the existence and uniqueness of the soft-Bellman equilibrium and propose a nonlinear least-squares algorithm to compute such an equilibrium in the forward problem. We then solve the inverse game problem of inferring the players' reward parameters from observed state-action trajectories via a projected-gradient algorithm. Experiments in a predator-prey OpenAI Gym environment show that the reward parameters inferred by the proposed algorithm outperform those inferred by a baseline algorithm: they reduce the Kullback-Leibler divergence between the equilibrium policies and observed policies by at least two orders of magnitude.