← 研踪 PaperTrailarXiv:2611.26252 · v1
The Stability Gap in Continual Reinforcement Learning
首次提交 2026-11-19 23:26
版本更新 2026-11-19 23:26
雷达观测 2026-11-20 14:52
摘要
Continual reinforcement learning (CRL) aims to enable agents to adapt to changing tasks while preserving previously acquired capabilities. Evaluation often relies on final aggregate metrics that can hide temporary performance degradation during adaptation. This work introduces a simple CRL benchmark that varies observation representations while preserving environment dynamics and rewards, supporting proof-of-concept demonstrations, inexpensive experimentation, and easily reproducible sanity checks. Using this benchmark, we demonstrate the stability gap in CRL: a substantial drop in prior-task performance after the introduction of new tasks, followed by partial or complete recovery. This pattern occurs despite the agent having complete access to past tasks.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-19 23:26 · 最新观测版本
The Stability Gap in Continual Reinforcement Learning
Continual reinforcement learning (CRL) aims to enable agents to adapt to changing tasks while preserving previously acquired capabilities. Evaluation often relies on final aggregate metrics that can hide temporary performance degradation during adaptation. This work introduces a simple CRL benchmark that varies observation representations while preserving environment dynamics and rewards, supporting proof-of-concept demonstrations, inexpensive experimentation, and easily reproducible sanity checks. Using this benchmark, we demonstrate the stability gap in CRL: a substantial drop in prior-task performance after the introduction of new tasks, followed by partial or complete recovery. This pattern occurs despite the agent having complete access to past tasks.
查看此版本解读 →查看此版本原文 ↗