In this paper, we study distributionally robust offline reinforcement learning (robust offline RL), which seeks to find an optimal policy purely from an offline dataset that can perform well in perturbed environments. In specific, we propose a generic algorithm framework called Doubly Pessimistic Model-based Policy Optimization ($P^2MPO$), which features a novel combination of a flexible model estimation subroutine and a doubly pessimistic policy optimization step. Notably, the double pessimism principle is crucial to overcome the distributional shifts incurred by (i) the mismatch between the behavior policy and the target policies; and (ii) the perturbation of the nominal model. Under certain accuracy conditions on the model estimation subroutine, we prove that $P^2MPO$ is sample-efficient with robust partial coverage data, which only requires the offline data to have good coverage of the distributions induced by the optimal robust policy and the perturbed models around the nominal model. By tailoring specific model estimation subroutines for concrete examples of RMDPs, including tabular RMDPs, factored RMDPs, kernel and neural RMDPs, we prove that $P^2MPO$ enjoys a $\tilde{\mathcal{O}}(n^{-1/2})$ convergence rate, where $n$ is the dataset size. We highlight that all these examples, except tabular RMDPs, are first identified and proven tractable by this work. Furthermore, we continue our study of robust offline RL in the robust Markov games (RMGs). By extending the double pessimism principle identified for single-agent RMDPs, we propose another algorithm framework that can efficiently find the robust Nash equilibria among players using only robust unilateral (partial) coverage data. To our best knowledge, this work proposes the first general learning principle -- double pessimism -- for robust offline RL and shows that it is provably efficient with general function approximation.

Clinical event sequences consist of hundreds of clinical events that represent records of patient care in time. Developing accurate predictive models of such sequences is of a great importance for supporting a variety of models for interpreting/classifying the current patient condition, or predicting adverse clinical events and outcomes, all aimed to improve patient care. One important challenge of learning predictive models of clinical sequences is their patient-specific variability. Based on underlying clinical conditions, each patient's sequence may consist of different sets of clinical events (observations, lab results, medications, procedures). Hence, simple population-wide models learned from event sequences for many different patients may not accurately predict patient-specific dynamics of event sequences and their differences. To address the problem, we propose and investigate multiple new event sequence prediction models and methods that let us better adjust the prediction for individual patients and their specific conditions. The methods developed in this work pursue refinement of population-wide models to subpopulations, self-adaptation, and a meta-level model switching that is able to adaptively select the model with the best chance to support the immediate prediction. We analyze and test the performance of these models on clinical event sequences of patients in MIMIC-III database.

Robust coordination skills enable agents to operate cohesively in shared environments, together towards a common goal and, ideally, individually without hindering each other's progress. To this end, this paper presents Coordinated QMIX (CoMIX), a novel training framework for decentralized agents that enables emergent coordination through flexible policies, allowing at the same time independent decision-making at individual level. CoMIX models selfish and collaborative behavior as incremental steps in each agent's decision process. This allows agents to dynamically adapt their behavior to different situations balancing independence and collaboration. Experiments using a variety of simulation environments demonstrate that CoMIX outperforms baselines on collaborative tasks. The results validate our incremental policy approach as effective technique for improving coordination in multi-agent systems.

Multi-agent interactions, such as communication, teaching, and bluffing, often rely on higher-order social inference, i.e., understanding how others infer oneself. Such intricate reasoning can be effectively modeled through nested multi-agent reasoning. Nonetheless, the computational complexity escalates exponentially with each level of reasoning, posing a significant challenge. However, humans effortlessly perform complex social inferences as part of their daily lives. To bridge the gap between human-like inference capabilities and computational limitations, we propose a novel approach: leveraging neural networks to amortize high-order social inference, thereby expediting nested multi-agent reasoning. We evaluate our method in two challenging multi-agent interaction domains. The experimental results demonstrate that our method is computationally efficient while exhibiting minimal degradation in accuracy.

Online decision making plays a crucial role in numerous real-world applications. In many scenarios, the decision is made based on performing a sequence of tests on the incoming data points. However, performing all tests can be expensive and is not always possible. In this paper, we provide a novel formulation of the online decision making problem based on combinatorial multi-armed bandits and take the cost of performing tests into account. Based on this formulation, we provide a new framework for cost-efficient online decision making which can utilize posterior sampling or BayesUCB for exploration. We provide a rigorous theoretical analysis for our framework and present various experimental results that demonstrate its applicability to real-world problems.

Recent advances in Unmanned Aerial Vehicles (UAVs) have resulted in their quick adoption for wide a range of civilian applications, including precision agriculture, biosecurity, disaster monitoring and surveillance. UAVs offer low-cost platforms with flexible hardware configurations, as well as an increasing number of autonomous capabilities, including take-off, landing, object tracking and obstacle avoidance. However, little attention has been paid to how UAVs deal with object detection uncertainties caused by false readings from vision-based detectors, data noise, vibrations, and occlusion. In most situations, the relevance and understanding of these detections are delegated to human operators, as many UAVs have limited cognition power to interact autonomously with the environment. This paper presents a framework for autonomous navigation under uncertainty in outdoor scenarios for small UAVs using a probabilistic-based motion planner. The framework is evaluated with real flight tests using a sub 2 kg quadrotor UAV and illustrated in victim finding Search and Rescue (SAR) case study in a forest/bushland. The navigation problem is modelled using a Partially Observable Markov Decision Process (POMDP), and solved in real time onboard the small UAV using Augmented Belief Trees (ABT) and the TAPIR toolkit. Results from experiments using colour and thermal imagery show that the proposed motion planner provides accurate victim localisation coordinates, as the UAV has the flexibility to interact with the environment and obtain clearer visualisations of any potential victims compared to the baseline motion planner. Incorporating this system allows optimised UAV surveillance operations by diminishing false positive readings from vision-based object detectors.

Q-learning has become an important part of the reinforcement learning toolkit since its introduction in the dissertation of Chris Watkins in the 1980s. The purpose of this paper is in part a tutorial on stochastic approximation and Q-learning, providing details regarding the INFORMS APS inaugural Applied Probability Trust Plenary Lecture, presented in Nancy France, June 2023. The paper also presents new approaches to ensure stability and potentially accelerated convergence for these algorithms, and stochastic approximation in other settings. Two contributions are entirely new: 1. Stability of Q-learning with linear function approximation has been an open topic for research for over three decades. It is shown that with appropriate optimistic training in the form of a modified Gibbs policy, there exists a solution to the projected Bellman equation, and the algorithm is stable (in terms of bounded parameter estimates). Convergence remains one of many open topics for research. 2. The new Zap Zero algorithm is designed to approximate the Newton-Raphson flow without matrix inversion. It is stable and convergent under mild assumptions on the mean flow vector field for the algorithm, and compatible statistical assumption on an underlying Markov chain. The algorithm is a general approach to stochastic approximation which in particular applies to Q-learning with "oblivious" training even with non-linear function approximation.

Navigating safely and efficiently in dense and heterogeneous traffic scenarios is challenging for autonomous vehicles (AVs) due to their inability to infer the behaviors or intentions of nearby drivers. In this work, we introduce a distributed multi-agent reinforcement learning (MARL) algorithm that can predict trajectories and intents in dense and heterogeneous traffic scenarios. Our approach for intent-aware planning, iPLAN, allows agents to infer nearby drivers' intents solely from their local observations. We model two distinct incentives for agents' strategies: Behavioral Incentive for high-level decision-making based on their driving behavior or personality and Instant Incentive for motion planning for collision avoidance based on the current traffic state. Our approach enables agents to infer their opponents' behavior incentives and integrate this inferred information into their decision-making and motion-planning processes. We perform experiments on two simulation environments, Non-Cooperative Navigation and Heterogeneous Highway. In Heterogeneous Highway, results show that, compared with centralized training decentralized execution (CTDE) MARL baselines such as QMIX and MAPPO, our method yields a 4.3% and 38.4% higher episodic reward in mild and chaotic traffic, with 48.1% higher success rate and 80.6% longer survival time in chaotic traffic. We also compare with a decentralized training decentralized execution (DTDE) baseline IPPO and demonstrate a higher episodic reward of 12.7% and 6.3% in mild traffic and chaotic traffic, 25.3% higher success rate, and 13.7% longer survival time.

We consider Linear Stochastic Approximation (LSA) with a constant stepsize and Markovian data. Viewing the joint process of the data and LSA iterate as a time-homogeneous Markov chain, we prove its convergence to a unique limiting and stationary distribution in Wasserstein distance and establish non-asymptotic, geometric convergence rates. Furthermore, we show that the bias vector of this limit admits an infinite series expansion with respect to the stepsize. Consequently, the bias is proportional to the stepsize up to higher order terms. This result stands in contrast with LSA under i.i.d. data, for which the bias vanishes. In the reversible chain setting, we provide a general characterization of the relationship between the bias and the mixing time of the Markovian data, establishing that they are roughly proportional to each other. While Polyak-Ruppert tail-averaging reduces the variance of the LSA iterates, it does not affect the bias. The above characterization allows us to show that the bias can be reduced using Richardson-Romberg extrapolation with $m\ge 2$ stepsizes, which eliminates the $m-1$ leading terms in the bias expansion. This extrapolation scheme leads to an exponentially smaller bias and an improved mean squared error, both in theory and empirically. Our results immediately apply to the Temporal Difference learning algorithm with linear function approximation, Markovian data, and constant stepsizes.

Cooperative Multi-agent Reinforcement Learning (MARL) algorithms with Zero-Shot Coordination (ZSC) have gained significant attention in recent years. ZSC refers to the ability of agents to coordinate zero-shot (without additional interaction experience) with independently trained agents. While ZSC is crucial for cooperative MARL agents, it might not be possible for complex tasks and changing environments. Agents also need to adapt and improve their performance with minimal interaction with other agents. In this work, we show empirically that state-of-the-art ZSC algorithms have poor performance when paired with agents trained with different learning methods, and they require millions of interaction samples to adapt to these new partners. To investigate this issue, we formally defined a framework based on a popular cooperative multi-agent game called Hanabi to evaluate the adaptability of MARL methods. In particular, we created a diverse set of pre-trained agents and defined a new metric called adaptation regret that measures the agent's ability to efficiently adapt and improve its coordination performance when paired with some held-out pool of partners on top of its ZSC performance. After evaluating several SOTA algorithms using our framework, our experiments reveal that naive Independent Q-Learning (IQL) agents in most cases adapt as quickly as the SOTA ZSC algorithm Off-Belief Learning (OBL). This finding raises an interesting research question: How to design MARL algorithms with high ZSC performance and capability of fast adaptation to unseen partners. As a first step, we studied the role of different hyper-parameters and design choices on the adaptability of current MARL algorithms. Our experiments show that two categories of hyper-parameters controlling the training data diversity and optimization process have a significant impact on the adaptability of Hanabi agents.