Trajectory Data Suffices for Statistically Efficient Learning in Offline RL with Linear $q^π$-Realizability and Concentrability

• URL: http://arxiv.org/abs/2405.16809v1
• Date: Mon, 27 May 2024 03:59:13 GMT
• Title: Trajectory Data Suffices for Statistically Efficient Learning in Offline RL with Linear $q^π$-Realizability and Concentrability
• Authors: Volodymyr Tkachuk, Gellért Weisz, Csaba Szepesvári,
• Abstract summary: offline reinforcement learning (RL) in $H$-horizon Markov decision processes (MDPs) We prove that with trajectory data, a dataset of size $textpoly(d,H,C_textconc)/epsilon2$ is sufficient for deriving an $epsilon$-optimal policy.
• Score: 34.51093353030245
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We consider offline reinforcement learning (RL) in $H$-horizon Markov decision processes (MDPs) under the linear $q^\pi$-realizability assumption, where the action-value function of every policy is linear with respect to a given $d$-dimensional feature function. The hope in this setting is that learning a good policy will be possible without requiring a sample size that scales with the number of states in the MDP. Foster et al. [2021] have shown this to be impossible even under $\textit{concentrability}$, a data coverage assumption where a coefficient $C_\text{conc}$ bounds the extent to which the state-action distribution of any policy can veer off the data distribution. However, the data in this previous work was in the form of a sequence of individual transitions. This leaves open the question of whether the negative result mentioned could be overcome if the data was composed of sequences of full trajectories. In this work we answer this question positively by proving that with trajectory data, a dataset of size $\text{poly}(d,H,C_\text{conc})/\epsilon^2$ is sufficient for deriving an $\epsilon$-optimal policy, regardless of the size of the state space. The main tool that makes this result possible is due to Weisz et al. [2023], who demonstrate that linear MDPs can be used to approximate linearly $q^\pi$-realizable MDPs. The connection to trajectory data is that the linear MDP approximation relies on "skipping" over certain states. The associated estimation problems are thus easy when working with trajectory data, while they remain nontrivial when working with individual transitions. The question of computational efficiency under our assumptions remains open.

Related papers

• Online RL in Linearly $q^\pi$-Realizable MDPs Is as Easy as in Linear MDPs If You Learn What to Ignore [0.0]
  We consider online reinforcement learning in episodic Markov decision processes (MDPs) We show that the difference between the two classes is the presence of states in linearly $qpi$-realizable MDPs. We derive a novel (computationally inefficient) learning algorithm for linearly $qpi$-realizable MDPs.
  arXiv  Detail & Related papers  (2023-10-11T18:50:25Z)
• On Instance-Dependent Bounds for Offline Reinforcement Learning with Linear Function Approximation [80.86358123230757]
  We present an algorithm called Bootstrapped and Constrained Pessimistic Value Iteration (BCP-VI) Under a partial data coverage assumption, BCP-VI yields a fast rate of $tildemathcalO(frac1K)$ for offline RL when there is a positive gap in the optimal Q-value functions. These are the first $tildemathcalO(frac1K)$ bound and absolute zero sub-optimality bound respectively for offline RL with linear function approximation from adaptive data.
  arXiv  Detail & Related papers  (2022-11-23T18:50:44Z)
• Provably Efficient Offline Reinforcement Learning with Trajectory-Wise Reward [66.81579829897392]
  We propose a novel offline reinforcement learning algorithm called Pessimistic vAlue iteRaTion with rEward Decomposition (PARTED) PARTED decomposes the trajectory return into per-step proxy rewards via least-squares-based reward redistribution, and then performs pessimistic value based on the learned proxy reward. To the best of our knowledge, PARTED is the first offline RL algorithm that is provably efficient in general MDP with trajectory-wise reward.
  arXiv  Detail & Related papers  (2022-06-13T19:11:22Z)
• Pessimism in the Face of Confounders: Provably Efficient Offline Reinforcement Learning in Partially Observable Markov Decision Processes [99.26864533035454]
  We study offline reinforcement learning (RL) in partially observable Markov decision processes. We propose the underlineProxy variable underlinePessimistic underlinePolicy underlineOptimization (textttP3O) algorithm. textttP3O is the first provably efficient offline RL algorithm for POMDPs with a confounded dataset.
  arXiv  Detail & Related papers  (2022-05-26T19:13:55Z)
• Settling the Sample Complexity of Model-Based Offline Reinforcement Learning [50.5790774201146]
  offline reinforcement learning (RL) learns using pre-collected data without further exploration. Prior algorithms or analyses either suffer from suboptimal sample complexities or incur high burn-in cost to reach sample optimality. We demonstrate that the model-based (or "plug-in") approach achieves minimax-optimal sample complexity without burn-in cost.
  arXiv  Detail & Related papers  (2022-04-11T17:26:19Z)
• Reward-Free RL is No Harder Than Reward-Aware RL in Linear Markov Decision Processes [61.11090361892306]
  Reward-free reinforcement learning (RL) considers the setting where the agent does not have access to a reward function during exploration. We show that this separation does not exist in the setting of linear MDPs. We develop a computationally efficient algorithm for reward-free RL in a $d$-dimensional linear MDP.
  arXiv  Detail & Related papers  (2022-01-26T22:09:59Z)
• Towards Instance-Optimal Offline Reinforcement Learning with Pessimism [34.54294677335518]
  We study the offline reinforcement learning (offline RL) problem, where the goal is to learn a reward-maximizing policy in an unknown Markov Decision Process (MDP) In this work, we analyze the Adaptive Pessimistic Value Iteration (APVI) algorithm and derive the suboptimality upper bound that nearly matches [ Oleft(sum_h=1Hsum_s_h,a_hdpistar_h(s_h,a_h)sqrtfracmathrmmathrmVar_
  arXiv  Detail & Related papers  (2021-10-17T01:21:52Z)
• Reward-Free Model-Based Reinforcement Learning with Linear Function Approximation [92.99933928528797]
  We study the model-based reward-free reinforcement learning with linear function approximation for episodic Markov decision processes (MDPs) In the planning phase, the agent is given a specific reward function and uses samples collected from the exploration phase to learn a good policy. We show that to obtain an $epsilon$-optimal policy for arbitrary reward function, UCRL-RFE needs to sample at most $tilde O(H4d(H + d)epsilon-2)$ episodes.
  arXiv  Detail & Related papers  (2021-10-12T23:03:58Z)
• On Query-efficient Planning in MDPs under Linear Realizability of the Optimal State-value Function [14.205660708980988]
  We consider the problem of local planning in fixed-horizon Markov Decision Processes (MDPs) with a generative model. A recent lower bound established that the related problem when the action-value function of the optimal policy is linearly realizable requires an exponential number of queries. In this work, we establish that poly$(H, d)$ learning is possible (with state value function realizability) whenever the action set is small.
  arXiv  Detail & Related papers  (2021-02-03T13:23:15Z)
• Online Robust Regression via SGD on the l1 loss [19.087335681007477]
  We consider the robust linear regression problem in the online setting where we have access to the data in a streaming manner. We show in this work that the descent on the $ell_O( 1 / (1 - eta)2 n )$ loss converges to the true parameter vector at a $tildeO( 1 / (1 - eta)2 n )$ rate which is independent of the values of the contaminated measurements.
  arXiv  Detail & Related papers  (2020-07-01T11:38:21Z)

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.