中文
相关论文

相关论文: Model-Free, Regret-Optimal Best Policy Identificat…

200 篇论文

In the optimization of dynamical systems, the variables typically have constraints. Such problems can be modeled as a constrained Markov Decision Process (CMDP). This paper considers a model-free approach to the problem, where the…

机器学习 · 计算机科学 2021-02-02 Qinbo Bai , Vaneet Aggarwal , Ather Gattami

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…

机器学习 · 计算机科学 2026-02-17 Orin Levy , Aviv Rosenberg , Alon Cohen , Yishay Mansour

In constrained Markov decision processes (CMDPs) with adversarial rewards and constraints, a well-known impossibility result prevents any algorithm from attaining both sublinear regret and sublinear constraint violation, when competing…

机器学习 · 计算机科学 2024-09-27 Francesco Emanuele Stradi , Anna Lunghi , Matteo Castiglioni , Alberto Marchesi , Nicola Gatti

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…

机器学习 · 计算机科学 2024-10-04 Francesco Emanuele Stradi , Matteo Castiglioni , Alberto Marchesi , Nicola Gatti

Constrained Markov decision processes (CMDPs) are a common way to model safety constraints in reinforcement learning. State-of-the-art methods for efficiently solving CMDPs are based on primal-dual algorithms. For these algorithms, all…

机器学习 · 计算机科学 2024-07-22 Adrian Müller , Pragnya Alatur , Volkan Cevher , Giorgia Ramponi , Niao He

We study \emph{online episodic Constrained Markov Decision Processes} (CMDPs) under both stochastic and adversarial constraints. We provide a novel algorithm whose guarantees greatly improve those of the state-of-the-art best-of-both-worlds…

We consider learning in an adversarial Markov Decision Process (MDP) where the loss functions can change arbitrarily over $K$ episodes and the state space can be arbitrarily large. We assume that the Q-function of any policy is linear in…

机器学习 · 计算机科学 2023-06-05 Yan Dai , Haipeng Luo , Chen-Yu Wei , Julian Zimmert

We study online learning in constrained Markov decision processes (CMDPs) in which rewards and constraints may be either stochastic or adversarial. In such settings, Stradi et al.(2024) proposed the first best-of-both-worlds algorithm able…

机器学习 · 计算机科学 2025-02-10 Francesco Emanuele Stradi , Anna Lunghi , Matteo Castiglioni , Alberto Marchesi , Nicola Gatti

Safety is a fundamental challenge in reinforcement learning (RL), particularly in real-world applications such as autonomous driving, robotics, and healthcare. To address this, Constrained Markov Decision Processes (CMDPs) are commonly used…

机器学习 · 计算机科学 2026-02-18 Chang Liu , Yunfan Li , Lin F. Yang

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…

机器学习 · 计算机科学 2026-03-31 Kihyun Yu , Seoungbin Bae , Dabeen Lee

We study online reinforcement learning in linear Markov decision processes with adversarial losses and bandit feedback, without prior knowledge on transitions or access to simulators. We introduce two algorithms that achieve improved regret…

机器学习 · 计算机科学 2023-10-19 Haolin Liu , Chen-Yu Wei , Julian Zimmert

We consider online learning for episodic stochastically constrained Markov decision processes (CMDPs), which plays a central role in ensuring the safety of reinforcement learning. Here the loss function can vary arbitrarily across the…

机器学习 · 计算机科学 2021-10-19 Shuang Qiu , Xiaohan Wei , Zhuoran Yang , Jieping Ye , Zhaoran Wang

A crucial problem in reinforcement learning is learning the optimal policy. We study this in tabular infinite-horizon discounted Markov decision processes under the online setting. The existing algorithms either fail to achieve regret…

机器学习 · 计算机科学 2023-12-13 Xiang Ji , Gen Li

We study online learning in constrained Markov decision processes (CMDPs) with adversarial losses and stochastic hard constraints, under bandit feedback. We consider three scenarios. In the first one, we address general CMDPs, where we…

机器学习 · 计算机科学 2025-02-10 Francesco Emanuele Stradi , Matteo Castiglioni , Alberto Marchesi , Nicola Gatti

We consider the problem of learning in adversarial Markov decision processes [MDPs] with an oblivious adversary in a full-information setting. The agent interacts with an environment during $T$ episodes, each of which consists of $H$…

机器学习 · 计算机科学 2025-03-06 Daniil Tiapkin , Evgenii Chzhen , Gilles Stoltz

Policy regret is a well established notion of measuring the performance of an online learning algorithm against an adaptive adversary. We study restrictions on the adversary that enable efficient minimization of the \emph{complete policy…

机器学习 · 统计学 2022-04-26 Dhruv Malik , Yuanzhi Li , Aarti Singh

We introduce two new no-regret algorithms for the stochastic shortest path (SSP) problem with a linear MDP that significantly improve over the only existing results of (Vial et al., 2021). Our first algorithm is computationally efficient…

机器学习 · 计算机科学 2021-12-21 Liyu Chen , Rahul Jain , Haipeng Luo

We present the OMG-CMDP! algorithm for regret minimization in adversarial Contextual MDPs. The algorithm operates under the minimal assumptions of realizable function class and access to online least squares and log loss regression oracles.…

机器学习 · 计算机科学 2023-08-15 Orin Levy , Alon Cohen , Asaf Cassel , Yishay Mansour

We study the problem of online learning in predictive control of an unknown linear dynamical system with time varying cost functions which are unknown apriori. Specifically, we study the online learning problem where the control algorithm…

机器学习 · 计算机科学 2022-11-01 Deepan Muthirayan , Jianjun Yuan , Dileep Kalathil , Pramod P. Khargonekar

Online safe reinforcement learning (RL) plays a key role in dynamic environments, with applications in autonomous driving, robotics, and cybersecurity. The objective is to learn optimal policies that maximize rewards while satisfying safety…

机器学习 · 计算机科学 2025-06-03 Jiahui Zhu , Kihyun Yu , Dabeen Lee , Xin Liu , Honghao Wei
‹ 上一页 1 2 3 10 下一页 ›