English
Related papers

Related papers: Policy Optimization in Adversarial MDPs: Improved …

200 papers

We introduce \texttt{OPO-CMDP}, the first policy optimization algorithm for stochastic Contextual Markov Decision Process (CMDPs) under general offline function approximation. Our approach achieves a high probability regret bound of…

Machine Learning · Computer Science 2026-02-17 Orin Levy , Aviv Rosenberg , Alon Cohen , Yishay Mansour

We consider an online learning problem where the learner interacts with a Markov decision process in a sequence of episodes, where the reward function is allowed to change between episodes in an adversarial manner and the learner only gets…

Machine Learning · Computer Science 2021-06-15 Gergely Neu , Julia Olkhovskaya

We present the first finite time global convergence analysis of policy gradient in the context of infinite horizon average reward Markov decision processes (MDPs). Specifically, we focus on ergodic tabular MDPs with finite state and action…

Machine Learning · Computer Science 2024-03-12 Navdeep Kumar , Yashaswini Murthy , Itai Shufaro , Kfir Y. Levy , R. Srikant , Shie Mannor

We consider the trade-off problem between exploration and exploitation under finite discounted Markov Decision Process, where the state transition matrix of the underlying environment stays unknown. We propose a double Thompson sampling…

Machine Learning · Computer Science 2022-03-01 Shuqing Shi , Xiaobin Wang , Zhiyou Yang , Fan Zhang , Hong Qu

Reinforcement learning typically assumes that agents observe feedback for their actions immediately, but in many real-world applications (like recommendation systems) feedback is observed in delay. This paper studies online learning in…

Machine Learning · Computer Science 2021-12-16 Tal Lancewicki , Aviv Rosenberg , Yishay Mansour

We study online finite-horizon Markov Decision Processes with adversarially changing loss and aggregate bandit feedback (a.k.a full-bandit). Under this type of feedback, the agent observes only the total loss incurred over the entire…

Machine Learning · Computer Science 2025-02-07 Tal Lancewicki , Yishay Mansour

A standard assumption in Reinforcement Learning is that the agent observes every visited state-action pair in the associated Markov Decision Process (MDP), along with the per-step rewards. Strong theoretical results are known in this…

Machine Learning · Computer Science 2026-02-03 Zhengjia Zhuo , Anupam Gupta , Viswanath Nagarajan

We study online learning in \emph{constrained MDPs} (CMDPs), focusing on the goal of attaining sublinear strong regret and strong cumulative constraint violation. Differently from their standard (weak) counterparts, these metrics do not…

Machine Learning · Computer Science 2024-10-04 Francesco Emanuele Stradi , Matteo Castiglioni , Alberto Marchesi , Nicola Gatti

We introduce and analyse two algorithms for exploration-exploitation in discrete and continuous Markov Decision Processes (MDPs) based on exploration bonuses. SCAL$^+$ is a variant of SCAL (Fruit et al., 2018) that performs efficient…

Machine Learning · Computer Science 2018-12-12 Jian Qian , Ronan Fruit , Matteo Pirotta , Alessandro Lazaric

We consider an agent interacting with an environment in a single stream of actions, observations, and rewards, with no reset. This process is not assumed to be a Markov Decision Process (MDP). Rather, the agent has several representations…

Machine Learning · Computer Science 2013-03-19 Odalric-Ambrym Maillard , Phuong Nguyen , Ronald Ortner , Daniil Ryabko

We study safe reinforcement learning in finite-horizon linear mixture constrained Markov decision processes (CMDPs) with adversarial rewards under full-information feedback and an unknown transition kernel. We propose a primal-dual policy…

Machine Learning · Computer Science 2026-03-31 Kihyun Yu , Seoungbin Bae , Dabeen Lee

We study the problem of infinite-horizon average-reward reinforcement learning with linear Markov decision processes (MDPs). The associated Bellman operator of the problem not being a contraction makes the algorithm design challenging.…

Machine Learning · Statistics 2025-03-12 Kihyuk Hong , Woojin Chae , Yufan Zhang , Dabeen Lee , Ambuj Tewari

Policy Optimization (PO) is a widely used approach to address continuous control tasks. In this paper, we introduce the notion of mediator feedback that frames PO as an online learning problem over the policy space. The additional available…

Machine Learning · Computer Science 2020-12-16 Alberto Maria Metelli , Matteo Papini , Pierluca D'Oro , Marcello Restelli

We study regret minimization in online episodic linear Markov Decision Processes, and obtain rate-optimal $\widetilde O (\sqrt K)$ regret where $K$ denotes the number of episodes. Our work is the first to establish the optimal (w.r.t.~$K$)…

Machine Learning · Computer Science 2024-05-17 Uri Sherman , Alon Cohen , Tomer Koren , Yishay Mansour

The standard assumption in reinforcement learning (RL) is that agents observe feedback for their actions immediately. However, in practice feedback is often observed in delay. This paper studies online learning in episodic Markov decision…

Machine Learning · Computer Science 2023-01-24 Tiancheng Jin , Tal Lancewicki , Haipeng Luo , Yishay Mansour , Aviv Rosenberg

While single-agent policy optimization in a fixed environment has attracted a lot of research attention recently in the reinforcement learning community, much less is known theoretically when there are multiple agents playing in a…

Machine Learning · Computer Science 2022-07-27 Shuang Qiu , Xiaohan Wei , Jieping Ye , Zhaoran Wang , Zhuoran Yang

We study minimax optimal reinforcement learning in episodic factored Markov decision processes (FMDPs), which are MDPs with conditionally independent transition components. Assuming the factorization is known, we propose two model-based…

Machine Learning · Computer Science 2020-06-25 Yi Tian , Jian Qian , Suvrit Sra

This work studies online episodic tabular Markov decision processes (MDPs) with known transitions and develops best-of-both-worlds algorithms that achieve refined data-dependent regret bounds in the adversarial regime and variance-dependent…

Machine Learning · Computer Science 2026-02-03 Mingyi Li , Taira Tsuchiya , Kenji Yamanishi

Policy Optimization (PO) methods are among the most popular Reinforcement Learning (RL) algorithms in practice. Recently, Sherman et al. [2023a] proposed a PO-based algorithm with rate-optimal regret guarantees under the linear Markov…

Machine Learning · Computer Science 2024-07-04 Asaf Cassel , Aviv Rosenberg

In this work, we study algorithms for learning in infinite-horizon undiscounted Markov decision processes (MDPs) with function approximation. We first show that the regret analysis of the Politex algorithm (a version of regularized policy…

Machine Learning · Computer Science 2021-02-26 Nevena Lazic , Dong Yin , Yasin Abbasi-Yadkori , Csaba Szepesvari