The Power of Resets in Online Reinforcement Learning

• URL: http://arxiv.org/abs/2404.15417v2
• Date: Fri, 26 Apr 2024 14:31:09 GMT
• Title: The Power of Resets in Online Reinforcement Learning
• Authors: Zakaria Mhammedi, Dylan J. Foster, Alexander Rakhlin,
• Abstract summary: We explore the power of simulators through online reinforcement learning with local simulator access (or, local planning) We show that MDPs with low coverability can be learned in a sample-efficient fashion with only $Qstar$-realizability. We show that the notorious Exogenous Block MDP problem is tractable under local simulator access.
• Score: 73.64852266145387
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Simulators are a pervasive tool in reinforcement learning, but most existing algorithms cannot efficiently exploit simulator access -- particularly in high-dimensional domains that require general function approximation. We explore the power of simulators through online reinforcement learning with {local simulator access} (or, local planning), an RL protocol where the agent is allowed to reset to previously observed states and follow their dynamics during training. We use local simulator access to unlock new statistical guarantees that were previously out of reach: - We show that MDPs with low coverability (Xie et al. 2023) -- a general structural condition that subsumes Block MDPs and Low-Rank MDPs -- can be learned in a sample-efficient fashion with only $Q^{\star}$-realizability (realizability of the optimal state-value function); existing online RL algorithms require significantly stronger representation conditions. - As a consequence, we show that the notorious Exogenous Block MDP problem (Efroni et al. 2022) is tractable under local simulator access. The results above are achieved through a computationally inefficient algorithm. We complement them with a more computationally efficient algorithm, RVFS (Recursive Value Function Search), which achieves provable sample complexity guarantees under a strengthened statistical assumption known as pushforward coverability. RVFS can be viewed as a principled, provable counterpart to a successful empirical paradigm that combines recursive search (e.g., MCTS) with value function approximation.

Related papers

• Efficient Function Placement in Virtual Networks: An Online Learning Approach [7.206295719344847]
  We propose a model for the virtual function placement problem and several novel algorithms using ideas based on multi-armed bandits. We prove that these algorithms learn the optimal placement policy rapidly, and their regret grows at a rate at most $O( N M sqrtTln T )$ while respecting the feasibility constraints with high probability.
  arXiv  Detail & Related papers  (2024-10-17T16:03:43Z)
• REBEL: Reinforcement Learning via Regressing Relative Rewards [59.68420022466047]
  We propose REBEL, a minimalist RL algorithm for the era of generative models. In theory, we prove that fundamental RL algorithms like Natural Policy Gradient can be seen as variants of REBEL. We find that REBEL provides a unified approach to language modeling and image generation with stronger or similar performance as PPO and DPO.
  arXiv  Detail & Related papers  (2024-04-25T17:20:45Z)
• A General Framework for Learning from Weak Supervision [93.89870459388185]
  This paper introduces a general framework for learning from weak supervision (GLWS) with a novel algorithm. Central to GLWS is an Expectation-Maximization (EM) formulation, adeptly accommodating various weak supervision sources. We also present an advanced algorithm that significantly simplifies the EM computational demands.
  arXiv  Detail & Related papers  (2024-02-02T21:48:50Z)
• The Effective Horizon Explains Deep RL Performance in Stochastic Environments [21.148001945560075]
  Reinforcement learning (RL) theory has largely focused on proving mini complexity sample bounds. We introduce a new RL algorithm, SQIRL, that iteratively learns a nearoptimal policy by exploring randomly to collect rollouts. We leverage SQIRL to derive instance-dependent sample complexity bounds for RL that are exponential only in an "effective horizon" look-ahead and on the complexity of the class used for approximation.
  arXiv  Detail & Related papers  (2023-12-13T18:58:56Z)
• Provably Efficient CVaR RL in Low-rank MDPs [58.58570425202862]
  We study risk-sensitive Reinforcement Learning (RL) We propose a novel Upper Confidence Bound (UCB) bonus-driven algorithm to balance interplay between exploration, exploitation, and representation learning in CVaR RL. We prove that our algorithm achieves a sample complexity of $epsilon$-optimal CVaR, where $H$ is the length of each episode, $A$ is the capacity of action space, and $d$ is the dimension of representations.
  arXiv  Detail & Related papers  (2023-11-20T17:44:40Z)
• Efficient Model-Free Exploration in Low-Rank MDPs [76.87340323826945]
  Low-Rank Markov Decision Processes offer a simple, yet expressive framework for RL with function approximation. Existing algorithms are either (1) computationally intractable, or (2) reliant upon restrictive statistical assumptions. We propose the first provably sample-efficient algorithm for exploration in Low-Rank MDPs.
  arXiv  Detail & Related papers  (2023-07-08T15:41:48Z)
• Sample Efficient Deep Reinforcement Learning via Local Planning [21.420851589712626]
  This work focuses on sample-efficient deep reinforcement learning (RL) with a simulator. We propose an algorithmic framework, named uncertainty-first local planning (UFLP), that takes advantage of this property. We demonstrate that this simple procedure can dramatically improve the sample cost of several baseline RL algorithms on difficult exploration tasks.
  arXiv  Detail & Related papers  (2023-01-29T23:17:26Z)
• Stabilizing Q-learning with Linear Architectures for Provably Efficient Learning [53.17258888552998]
  This work proposes an exploration variant of the basic $Q$-learning protocol with linear function approximation. We show that the performance of the algorithm degrades very gracefully under a novel and more permissive notion of approximation error.
  arXiv  Detail & Related papers  (2022-06-01T23:26:51Z)
• Efficient Local Planning with Linear Function Approximation [27.90696655434707]
  We study query and computationally efficient planning algorithms with linear function approximation and a simulator. We propose an algorithm named confident Monte Carlo least square policy iteration (Confident MC-LSPI) for this setting. One technical contribution of our work is the introduction of a novel proof technique that makes use of a virtual policy algorithm.
  arXiv  Detail & Related papers  (2021-08-12T04:56:33Z)

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.