mlsanityData to 7 Oct 2026

Probabilistic Performance Guarantees for Multi-Task Reinforcement Learning

arXiv 2602.02098 ↗First posted 2 Feb 2026cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Multi-task reinforcement learning trains generalist policies that can execute multiple tasks. While recent years have seen significant progress, existing approaches rarely provide formal performance guarantees, which are indispensable when deploying policies in safety-critical settings. We present an approach for computing high-confidence guarantees on the performance of a multi-task policy on tasks not seen during training. Concretely, we introduce a new generalisation bound that composes (i) per-task lower confidence bounds from finitely many rollouts with (ii) task-level generalisation from finitely many sampled tasks, yielding a high-confidence guarantee for new tasks drawn from the same arbitrary and unknown distribution. Across state-of-the-art multi-task RL methods, we show that the guarantees are theoretically sound and informative at realistic sample sizes.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 15 of its 39 references that are in the map. First posted on arXiv 2 Feb 2026.

Builds on

15 of 39 references.

References · 15

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  3. 2016RL2: Fast Reinforcement Learning via Slow Reinforcement Learningcited by 490
  4. 2021Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulationcited by 375
  5. 2019Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variablescited by 372
  6. 2016Learning to reinforcement learncited by 369
  7. 2020dm_control: Software and Tasks for Continuous Controlcited by 368
  8. 2017Distral: Robust Multitask Reinforcement Learningcited by 231
  9. 2015Actor-Mimic: Deep Multitask and Transfer Reinforcement Learningcited by 176
  10. 2017Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learningcited by 97
  11. 2021LTL2Action: Generalizing LTL Instructions for Multi-Task RLcited by 72
  12. 2024DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RLcited by 20
3 more
  1. 2023Train Hard, Fight Easy: Robust Meta Reinforcement Learningcited by 11
  2. 2024Certifiably Robust Policies for Uncertain Parametric Environmentscited by 6
  3. 2024Sound Statistical Model Checking for Probabilities and Expected Rewards (extended version)cited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2022On Practical Reinforcement Learning: Provable Robustness, Scalability, and Statistical Efficiency
  2. 2020Provably Good Batch Reinforcement Learning Without Great Exploration
  3. 2025Stabilizing Reinforcement Learning with LLMs: Formulation and Practices
  4. 2025Generalisation in Multitask Fitted Q-Iteration and Offline Q-learning
  5. 2024The Power of Active Multi-Task Learning in Reinforcement Learning from Human Feedback
  6. 2024Offline Multi-task Transfer RL with Representational Penalization
  7. 2019Multi-task Batch Reinforcement Learning with Metric Learning
  8. 2026Maximum Likelihood Reinforcement Learning