Probabilistic performance guarantees for multi-task reinforcement learning

Conference item

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.

Authors: Schnitzer, Y; Jackermeier, M; Abate, A; Parker, DA

• Publication status: Accepted
• Peer review status: Peer reviewed
• Host title: Proceedings of the 43rd International Conference on Machine Learning (PMLR 306)
• Acceptance date: 2026-04-30
• Event title: 43rd International Conference on Machine Learning (ICML'26)
• Event location: Seoul, South Korea
• Event website: https://icml.cc/
• Event start date: 2026-07-06
• Event end date: 2026-07-11
• Language: English