arXiv:2502.07978cs.LG2025-02综述被引 37
不更新参数就能学会新任务，这篇综述讲清了上下文强化学习的原理与进展。
A Survey of In-Context Reinforcement Learning
- 利用动作-观测历史作为上下文，无需反向传播更新参数
- 在多个新任务上实现零样本适应，性能接近传统强化学习
- 适合研究高效学习、少样本智能的学者参考
强化学习代理通常通过昂贵的反向传播来更新网络参数以优化策略。然而，一些代理可以通过仅依赖额外的上下文（如动作-观测历史）来解决新任务，而无需更新任何参数。本文综述了此类行为，即上下文强化学习的研究进展。
原文摘要 · Abstract (English)
Reinforcement learning (RL) agents typically optimize their policies by performing expensive backward passes to update their network parameters. However, some agents can solve new tasks without updating any parameters by simply conditioning on additional context such as their action-observation histories. This paper surveys work on such behavior, known as in-context reinforcement learning.
强化学习上下文学习零样本
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。