arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.11574cs.LG
收藏

面向强化学习的离线多任务表示学习

Offline Multitask Representation Learning for Reinforcement Learning

Mila, McGill University(米拉研究所,麦吉尔大学);Johns Hopkins University(约翰斯·霍普金斯大学);University of Florida(佛罗里达大学);Princeton University(普林斯顿大学)

Haque Ishfaq, Thanh Nguyen-Tang, Songtao Feng, Raman Arora, Mengdi Wang, Ming Yin, Doina Precup

AI总结:

针对强化学习中的离线多任务表示学习问题,提出MORL算法,理论验证了利用上游离线任务所学表示相较直接学习低秩模型表示在多类下游RL场景中更具优势。

摘要:

我们研究强化学习(RL)中的离线多任务表示学习问题:学习者接收来自多个共享通用表示的不同任务的离线数据集,需学习该共享表示。我们从理论上分析了离线多任务低秩RL,并提出一种名为MORL的新型离线多任务表示学习算法。此外,我们还探究了无奖励、离线及在线场景下的下游RL任务——向智能体引入与上游离线任务共享同一表示的新任务。理论结果表明,相比直接学习低秩模型的表示,利用上游离线任务学到的表示更具优势。

英文摘要:

We study offline multitask representation learning in reinforcement learning (RL), where a learner is provided with an offline dataset from different tasks that share a common representation and is asked to learn the shared representation. We theoretically investigate offline multitask low-rank RL, and propose a new algorithm called MORL for offline multitask representation learning. Furthermore, we examine downstream RL in reward-free, offline and online scenarios, where a new task is introduced to the agent that shares the same representation as the upstream offline tasks. Our theoretical results demonstrate the benefits of using the learned representation from the upstream offline task instead of directly learning the representation of the low-rank model.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑