Pure Exploration for a Good Policy in Reinforcement Learning with Bandit Feedback

• URL: http://arxiv.org/abs/2605.23182v1
• Date: Fri, 22 May 2026 03:04:45 GMT
• Title: Pure Exploration for a Good Policy in Reinforcement Learning with Bandit Feedback
• Abstract summary: Best Policy Identification (BPI) seeks to identify a (near)-optimal policy with high confidence.<n>We formalize Good Policy Identification (GPI) under the fixed-confidence setting.<n>We propose a novel algorithm BEE-GPI, and derive theoretically-grounded upper bounds on its sample complexity for positive and negative instances.
• Score: 8.450904497835262
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Pure exploration in episodic Reinforcement Learning has primarily focused on Best Policy Identification (BPI), which seeks to identify a (near)-optimal policy with high confidence. Motivated by practical settings where a ``good enough'' policy suffices, we study an alternate objective of Good Policy Identification (GPI). For a given reward threshold $μ_0$, GPI only requires identifying a policy with expected reward in an episode at least $μ_0$ if such a policy exists (positive instance), or declaring None if no such policy exists (negative instance). We formalize GPI under the fixed-confidence setting. We require the output to be correct with probability $\geq 1-δ$, and seek to minimize the expected sample complexity, which is the expected number of episodes explored for the output. We propose a novel algorithm BEE-GPI, and derive theoretically-grounded upper bounds on its sample complexity for positive and negative instances. Notably, for positive instances, the coefficient of $\log 1/δ$ in our upper bound is $O(H^2/(V^* - μ_0)^2)$, where $H$ is the episode length and $V^*$ is the optimal expected reward in an episode. The coefficient does not depend on the action and state space sizes otherwise, in sharp contrast to the sample complexity in BPI. We further establish lower bound results to show the near-optimality of BEE-GPI and the necessity of the $1/(V^* -μ)^2$ term. Numerical experiments further validate the efficiency of our approach.

Related papers

• Computational Hardness of Reinforcement Learning with Partial $q^π$-Realizability [1.6328866317851185]
  This paper investigates the computational complexity of reinforcement learning in a novel linear function approximation regime, termed partial $qpi$-realizability.<n>We prove that learning an $epsilon$-optimal policy in this setting is computationally hard.<n>Our results mirror those in $q*$-realizability and suggest computational difficulty persists even when $Pi$ is expanded beyond the optimal policy.
  arXiv  Detail & Related papers  (2025-10-24T01:18:49Z)
• Distributionally Robust Policy Learning under Concept Drifts [33.44768994272614]
  This paper studies a more nuanced problem -- robust policy learning under the concept drift.<n>We first provide a doubly-robust estimator for evaluating the worst-case average reward of a given policy.<n>We then propose a learning algorithm that outputs the policy maximizing the estimated policy value within a given policy class.
  arXiv  Detail & Related papers  (2024-12-18T19:53:56Z)
• Estimating Optimal Policy Value in General Linear Contextual Bandits [50.008542459050155]
  In many bandit problems, the maximal reward achievable by a policy is often unknown in advance. We consider the problem of estimating the optimal policy value in the sublinear data regime before the optimal policy is even learnable. We present a more practical, computationally efficient algorithm that estimates a problem-dependent upper bound on $V*$.
  arXiv  Detail & Related papers  (2023-02-19T01:09:24Z)
• Reaching Goals is Hard: Settling the Sample Complexity of the Stochastic Shortest Path [106.37656068276902]
  We study the sample complexity of learning an $epsilon$-optimal policy in the Shortest Path (SSP) problem. We derive complexity bounds when the learner has access to a generative model. We show that there exists a worst-case SSP instance with $S$ states, $A$ actions, minimum cost $c_min$, and maximum expected cost of the optimal policy over all states $B_star$.
  arXiv  Detail & Related papers  (2022-10-10T18:34:32Z)
• Best Policy Identification in Linear MDPs [70.57916977441262]
  We investigate the problem of best identification in discounted linear Markov+Delta Decision in the fixed confidence setting under a generative model. The lower bound as the solution of an intricate non- optimization program can be used as the starting point to devise such algorithms.
  arXiv  Detail & Related papers  (2022-08-11T04:12:50Z)
• The Curse of Passive Data Collection in Batch Reinforcement Learning [82.6026077420886]
  In high stake applications, active experimentation may be considered too risky and thus data are often collected passively. While in simple cases, such as in bandits, passive and active data collection are similarly effective, the price of passive sampling can be much higher when collecting data from a system with controlled states.
  arXiv  Detail & Related papers  (2021-06-18T07:54:23Z)
• Robust Policy Gradient against Strong Data Corruption [30.910088777897045]
  We study the problem of robust reinforcement learning under adversarial corruption on both rewards and transitions. Our attack model assumes an textitadaptive adversary who can arbitrarily corrupt the reward and transition at every step within an episode. We develop a Filtered Policy Gradient algorithm that can tolerate even reward corruption and can find an $O(epsilon1/4)$-optimal policy.
  arXiv  Detail & Related papers  (2021-02-11T01:48:38Z)
• Adaptive Reward-Free Exploration [48.98199700043158]
  We show that our reward-free UCRL algorithm can be seen as a variant of an algorithm of Fiechter from 1994. We further investigate the relative complexities of reward-free exploration and best-policy identification.
  arXiv  Detail & Related papers  (2020-06-11T09:58:03Z)
• Provably Efficient Safe Exploration via Primal-Dual Policy Optimization [105.7510838453122]
  We study the Safe Reinforcement Learning (SRL) problem using the Constrained Markov Decision Process (CMDP) formulation. We present an provably efficient online policy optimization algorithm for CMDP with safe exploration in the function approximation setting.
  arXiv  Detail & Related papers  (2020-03-01T17:47:03Z)
• Reward-Free Exploration for Reinforcement Learning [82.3300753751066]
  We propose a new "reward-free RL" framework to isolate the challenges of exploration. We give an efficient algorithm that conducts $tildemathcalO(S2Amathrmpoly(H)/epsilon2)$ episodes of exploration. We also give a nearly-matching $Omega(S2AH2/epsilon2)$ lower bound, demonstrating the near-optimality of our algorithm in this setting.
  arXiv  Detail & Related papers  (2020-02-07T14:03:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.