Making decisions in the presence of a strategic opponent requires one to take into account the opponent's ability to actively mask its intended objective. To describe such strategic situations, we introduce the non-cooperative inverse reinforcement learning (N-CIRL) formalism. The N-CIRL formalism consists of two agents with completely misaligned objectives, where only one of the agents knows the true objective function.

This paper introduces a novel backup strategy for Monte-Carlo Tree Search (MCTS) designed for highly stochastic and partially observable Markov decision processes. We adopt a probabilistic approach, modeling both value and action-value nodes as Gaussian distributions. We introduce a novel backup operator that computes value nodes as the Wasserstein barycenter of their action-value children nodes; thus, propagating the uncertainty of the estimate across the tree to the root node. We study our novel backup operator when using a novel combination of $L^1$-Wasserstein barycenter with $\alpha$-divergence, by drawing a notable connection to the generalized mean backup operator. We complement our probabilistic backup operator with two sampling strategies, based on optimistic selection and Thompson sampling, obtaining our Wasserstein MCTS algorithm. We provide theoretical guarantees of asymptotic convergence to the optimal policy, and an empirical evaluation on several stochastic and partially observable environments, where our approach outperforms well-known related baselines.

Monte-Carlo Tree Search (MCTS) is a class of methods for solving complex decision-making problems through the synergy of Monte-Carlo planning and Reinforcement Learning (RL). The highly combinatorial nature of the problems commonly addressed by MCTS requires the use of efficient exploration strategies for navigating the planning tree and quickly convergent value backup methods. These crucial problems are particularly evident in recent advances that combine MCTS with deep neural networks for function approximation. In this work, we propose two methods for improving the convergence rate and exploration based on a newly introduced backup operator and entropy regularization. We provide strong theoretical guarantees to bound convergence rate, approximation error, and regret of our methods. Moreover, we introduce a mathematical framework based on the use of the $\alpha$-divergence for backup and exploration in MCTS. We show that this theoretical formulation unifies different approaches, including our newly introduced ones, under the same mathematical framework, allowing to obtain different methods by simply changing the value of $\alpha$. In practice, our unified perspective offers a flexible way to balance between exploration and exploitation by tuning the single $\alpha$ parameter according to the problem at hand. We validate our methods through a rigorous empirical study from basic toy problems to the complex Atari games, and including both MDP and POMDP problems.

Off-policy reinforcement learning (RL) holds the promise of sample-efficient learning of decision-making policies by leveraging past experience. However, in the offline RL setting -- where a fixed collection of interactions are provided and no further interactions are allowed -- it has been shown that standard off-policy RL methods can significantly underperform. Recently proposed methods aim to address this shortcoming by regularizing learned policies to remain close to the given dataset of interactions. However, these methods involve several configurable components such as learning a separate policy network on top of a behavior cloning actor, and explicitly constraining action spaces through clipping or reward penalties. Striving for simultaneous simplicity and performance, in this work we present a novel backup operator, Expected-Max Q-Learning (EMaQ), which naturally restricts learned policies to remain within the support of the offline dataset \emph{without any explicit regularization}, while retaining desirable theoretical properties such as contraction. We demonstrate that EMaQ is competitive with Soft Actor Critic (SAC) in online RL, and surpasses SAC in the deployment-efficient setting. In the offline RL setting -- the main focus of this work -- through EMaQ we are able to make important observations regarding key components of offline RL, and the nature of standard benchmark tasks. Lastly but importantly, we observe that EMaQ achieves state-of-the-art performance with fewer moving parts such as one less function approximation, making it a strong, yet easy to implement baseline for future work.

Making decisions in the presence of a strategic opponent requires one to take into account the opponent's ability to actively mask its intended objective. To describe such strategic situations, we introduce the non-cooperative inverse reinforcement learning (N-CIRL) formalism. The N-CIRL formalism consists of two agents with completely misaligned objectives, where only one of the agents knows the true objective function. Formally, we model the N-CIRL formalism as a zero-sum Markov game with one-sided incomplete information. Through interacting with the more informed player, the less informed player attempts to both infer, and act according to, the true objective function. As a result of the one-sided incomplete information, the multi-stage game can be decomposed into a sequence of single-stage games expressed by a recursive formula. Solving this recursive formula yields the value of the N-CIRL game and the more informed player's equilibrium strategy. Another recursive formula, constructed by forming an auxiliary game, termed the dual game, yields the less informed player's strategy. Building upon these two recursive formulas, we develop a computationally tractable algorithm to approximately solve for the equilibrium strategies. Finally, we demonstrate the benefits of our N-CIRL formalism over the existing multi-agent IRL formalism via extensive numerical simulation in a novel cyber security setting.

We consider Monte-Carlo Tree Search (MCTS) applied to Markov Decision Processes (MDPs) and Partially Observable MDPs (POMDPs), and the well-known Upper Confidence bound for Trees (UCT) algorithm. In UCT, a tree with nodes (states) and edges (actions) is incrementally built by the expansion of nodes, and the values of nodes are updated through a backup strategy based on the average value of child nodes. However, it has been shown that with enough samples the maximum operator yields more accurate node value estimates than averaging. Instead of settling for one of these value estimates, we go a step further proposing a novel backup strategy which uses the power mean operator, which computes a value between the average and maximum value. We call our new approach Power-UCT and argue how the use of the power mean operator helps to speed up the learning in MCTS. We theoretically analyze our method providing guarantees of convergence to the optimum. Moreover, we discuss a heuristic approach to balance the greediness of backups by tuning the power mean operator according to the number of visits to each node. Finally, we empirically demonstrate the effectiveness of our method in well-known MDP and POMDP benchmarks, showing significant improvement in performance and convergence speed w.r.t. UCT.

Driverless cars plying California's roads are about to get a little more driverless. Under new regulation passed by the state's Department of Motor Vehicles (DMV), self-driving cars will be permitted to use public roads without carrying a human who can take over if things go awry. The regulations still require people to be supervising the cars remotely. The news represents a major step for the burgeoning autonomous car industry and eases the path for California to continue playing a prominent role. Arizona has already cleared the way for cars without human operators, issuing a permit this year for Waymo - a unit of Alphabet Inc - to operate a commercial ride-hailing service using a fleet of driverless cars.

Monte-Carlo Tree Search (MCTS) algorithms estimate the value of MDP states based on rewards received by performing multiple random simulations. MCTS algorithms can use different strategies to aggregate these rewards and provide an estimation for the states’ values. The most common aggregation method is to store the mean reward of all simulations. Another common approach stores the best observed reward from each state. Both of these methods have complementary benefits and drawbacks. In this paper, we show that both of these methods are biased estimators for the real expected value of MDP states. We propose an hybrid approach that uses the best reward for states with low noise, and otherwise uses the mean. Experimental results on the Sailing MDP domain show that our method has a considerable advantage when the rewards are drawn from a noisy distribution.

Kernel-based reinforcement learning (KBRL) is a popular approach to learning non-parametric value function approximations. In this paper, we present structured KBRL, a paradigm for kernel-based RL that allows for modeling independencies in the transition and reward models of problems. Real-world problems often exhibit this structure and can be solved more efficiently when it is modeled. We make three contributions. First, we motivate our work, define a structured backup operator, and prove that it is a contraction. Second, we show how to evaluate our operator efficiently. Our analysis reveals that the fixed point of the operator is the optimal value function in a special factored MDP. Finally, we evaluate our method on a synthetic problem and compare it to two KBRL baselines. In most experiments, we learn better policies than the baselines from an order of magnitude less training data.