mlsanityData to 7 Oct 2026

An Information-Theoretic Analysis of OOD Generalization in Meta-Reinforcement Learning

arXiv 2510.23448 ↗First posted 27 Oct 2025cs.LG stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

In this work, we study out-of-distribution (OOD) generalization in meta-reinforcement learning from an information-theoretic perspective. We begin by establishing OOD generalization bounds for meta-supervised learning under two distinct distribution shift scenarios: standard distribution mismatch and a broad-to-narrow training setting. Building on this foundation, we formalize the generalization problem in meta-reinforcement learning and establish fine-grained generalization bounds that exploit the structure of Markov Decision Processes. Lastly, we analyze the generalization performance of a gradient-based meta-reinforcement learning algorithm.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 58 of its 80 references that are in the map. First posted on arXiv 27 Oct 2025.

Builds on

58 of 80 references.

References · 58

  1. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  2. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  3. 2020Meta-Learning in Neural Networks: A Surveycited by 944
  4. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 762
  5. 2019Fantastic Generalization Measures and Where to Find Themcited by 507
  6. 2017Information-theoretic analysis of generalization capability of learning algorithmscited by 299
  7. 2018Learning to Adapt in Dynamic, Real-World Environments Through Meta-Reinforcement Learningcited by 270
  8. 2020Few-Shot Learning via Learning the Representation, Provablycited by 213
  9. 2020Provable Meta-Learning of Linear Representationscited by 161
  10. 2018Meta-Gradient Reinforcement Learningcited by 156
  11. 2020Reasoning About Generalization via Conditional Mutual Informationcited by 145
  12. 2018Meta-Reinforcement Learning of Structured Exploration Strategiescited by 128
38 more
  1. 2019Information-Theoretic Generalization Bounds for SGLD via Data-Dependent Estimatescited by 118
  2. 2017Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environmentscited by 106
  3. 2023A Tutorial on Meta-Reinforcement Learningcited by 102
  4. 2020Sharpened Generalization Bounds based on Conditional Mutual Information and an Application to Noisy, Iterative Algorithmscited by 89
  5. 2020PACOH: Bayes-Optimal Meta-Learning with PAC-Guaranteescited by 85
  6. 2021Information-Theoretic Generalization Bounds for Stochastic Gradient Descentcited by 84
  7. 2018Generalization Error Bounds for Noisy, Iterative Algorithmscited by 79
  8. 2017Meta-Learning by Adjusting Priors Based on Extended PAC-Bayes Theorycited by 76
  9. 2019Improving Generalization in Meta Reinforcement Learning using Learned Objectivescited by 68
  10. 2018Chaining Mutual Information and Tightening Generalization Boundscited by 65
  11. 2023Generalization Bounds: Perspectives from Information Theory and PAC-Bayescited by 60
  12. 2020Meta-Gradient Reinforcement Learning with an Objective Discovered Onlinecited by 52
  13. 2021Information-theoretic generalization bounds for black-box learning algorithmscited by 49
  14. 2019Learning to Balance: Bayesian Meta-Learning for Imbalanced and Out-of-distribution Taskscited by 48
  15. 2019Generalization Error Bounds Via R'enyi-, f-Divergences and Maximal Leakagecited by 48
  16. 2021Generalization Bounds For Meta-Learning: An Information-Theoretic Analysiscited by 45
  17. 2018Some Considerations on Learning to Explore via Meta-Reinforcement Learningcited by 43
  18. 2021Bayesian decision-making under misspecified priors with applications to meta-learningcited by 43
  19. 2021Risk Bounds and Rademacher Complexity in Batch Reinforcement Learningcited by 42
  20. 2021Generalization of Model-Agnostic Meta-Learning Algorithms: Recurring and Unseen Taskscited by 36
  21. 2020Information-Theoretic Generalization Bounds for Meta-Learning and Applicationscited by 32
  22. 2020Individually Conditional Individual Mutual Information Bound on Generalization Errorcited by 30
  23. 2021Generalization Bounds for Meta-Learning via PAC-Bayes and Uniform Stabilitycited by 29
  24. 2021Towards a Unified Information-Theoretic Framework for Generalizationcited by 29
  25. 2020Information-theoretic analysis for transfer learningcited by 23
  26. 2025Langevin Soft Actor-Critic: Efficient Exploration through Uncertainty-Driven Critic Learningcited by 17
  27. 2021Generalization Bounds for Noisy Iterative Algorithms Using Properties of Additive Noise Channelscited by 15
  28. 2022Information-Theoretic Analysis of Unsupervised Domain Adaptationcited by 15
  29. 2021Information-Theoretic Bounds on the Moments of the Generalization Error of Learning Algorithmscited by 12
  30. 2021Improving Generalization in Meta-RL with Imaginary Tasks from Latent Dynamics Mixturecited by 12
  31. 2022Distributionally Adaptive Meta Reinforcement Learningcited by 11
  32. 2022Evaluated CMI Bounds for Meta Learning: Tightness and Expressivenesscited by 11
  33. 2023Metalearning with Very Few Samples Per Taskcited by 11
  34. 2024More Flexible PAC-Bayesian Meta-Learning by Learning Learning Algorithmscited by 10
  35. 2024Information-Theoretic Generalization Bounds for Deep Neural Networkscited by 10
  36. 2021Two Sides of Meta-Learning Evaluation: In vs. Out of Distributioncited by 9
  37. 2021Regularization Guarantees Generalization in Bayesian Reinforcement Learning through Algorithmic Stabilitycited by 7
  38. 2022Meta Reinforcement Learning with Finite Training Tasks -- a Density Estimation Approachcited by 6

50 most cited of 58 All 58 (JSON)

Built on by

Not cited yet.