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 2402.06402cs.LGcs.AI
收藏

分层Transformer是高效元强化学习器

Hierarchical Transformers are Efficient Meta-Reinforcement Learners

University of Freiburg(弗赖堡大学);University of Technology Nuremberg(纽伦堡工业大学)

Gresa Shala, André Biedenkapp, Josif Grabocka

AI总结:

本文提出分层Transformer元强化学习方法(HTrMRL),利用过去经验提升新任务适应性与泛化能力,在Meta-World基准上超越现有最先进方法。

摘要:

我们提出了用于元强化学习的分层Transformer(HTrMRL),这是一种强大的在线元强化学习方法。HTrMRL旨在解决使强化学习代理在未见过的任务中有效执行的挑战。我们展示了过去的经验如何作为丰富的信息来源,我们的模型能够有效地提炼并应用于新情境。我们学习到的算法能够超越先前的最先进水平,提供更高效的元训练,同时显著提升泛化能力。在Meta-World基准测试的多个模拟任务上获得的实验结果表明,与最先进方法相比,我们的方法在多种任务上的学习效率和适应性均有显著提升。我们的方法不仅增强了代理从有限数据中泛化的能力,还为构建更健壮和多功能的AI系统铺平了道路。

英文摘要:

We introduce Hierarchical Transformers for Meta-Reinforcement Learning (HTrMRL), a powerful online meta-reinforcement learning approach. HTrMRL aims to address the challenge of enabling reinforcement learning agents to perform effectively in previously unseen tasks. We demonstrate how past episodes serve as a rich source of information, which our model effectively distills and applies to new contexts. Our learned algorithm is capable of outperforming the previous state-of-the-art and provides more efficient meta-training while significantly improving generalization capabilities. Experimental results, obtained across various simulated tasks of the Meta-World Benchmark, indicate a significant improvement in learning efficiency and adaptability compared to the state-of-the-art on a variety of tasks. Our approach not only enhances the agent's ability to generalize from limited data but also paves the way for more robust and versatile AI systems.

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。

↑