cs.LGNov 16, 2026

Adaptive Memory for Online Reinforcement Learning under Sparse Model Changes

Authors: Sreejeet Maity, Sukanya Modak, Leonardo F. Toso

Organizations: Department of Electrical and Computer Engineering, North Carolina State University · Department of Electronics and Communication Engineering, RCC Institute of Information Technology · Department of Electrical Engineering, Columbia University

Abstract

We study the discounted reinforcement learning (RL) problem in a setting where the underlying environment changes between deployments. Each deployment provides HH observations from a fixed MDP. The reward function and transition kernel are fixed within each deployment. Across deployments, at most an ε>0ε>0 fraction of the reward and transition kernel entries may change. In this setting, learning each model to deploy from scratch may discard useful observations from past deployments, whereas pooling all past observations can bias the estimates of entries that have already changed. We then propose an algorithm that for each state and action pair, selects how much history to reuse to estimate the reward and probabilities of the next state. We start with observations from the current deployment and add earlier deployment observations one at a time, the most recent first. After each addition, we check whether a single average reward agrees with the estimates from all histories considered until that point. We apply the same check to the distribution of next states. For each estimate, we leverage the last history that passes these aforementioned checks. The optimal control policy is then computed on top of the fitted MDP. We characterize the average gap between the optimal return and the return of the learner's policies over a finite number of deployments. We demonstrate that when 0<∣S∣ε≤20< |S|ε\leq 2, with ∣S∣|S| being the cardinality of the state space, our algorithm needs only the observations from the most recent O(1/ε)O(1/ε) deployments, and achieves a long-run average gap of at most O~(ε/H)\widetilde{O}(\sqrt{ε/H}).

Figures & tables

Explore similar work

CardsList
  1. Data Deletion Can Help in Adaptive RL

    Apr 30, 2026Param Budhraja, Aditya Gangrade, Alex Olshevsky +1Non-Stationary RLReinforcement Learning

  2. From Static Policies to Adaptive Priors in Offline Reinforcement Learning

    Sep 26, 2026Tianwei Ni, Vineet Jain, Akash Karthikeyan +1Offline RL

  3. ATLAS: Adaptive Topological Learning with Abstract Successors for Continual Learning

    Aug 5, 2026R. Blake Lawlor, Daniel S. BrownNon-Stationary RLReinforcement Learning