arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2112.05908cs.LGcs.SYeess.SY
收藏

边缘的联邦强化学习

Federated Reinforcement Learning at the Edge

University of Oxford(牛津大学)

Konstantinos Gatsis

摘要:

现代网络物理架构使用从不同物理位置的系统收集的数据来学习适当的行为并适应不确定的环境。然而,一个重要的挑战出现,因为在网络系统的边缘通信交换是昂贵的,由于有限的资源。本文考虑了一种设置,多个代理需要有效地进行通信,以共同解决强化学习问题的时间序列数据收集在一个分布式的方式。这被设定为通过通信网络学习近似价值函数。提出了一种实现通信效率的算法,支持理论保证,实际实现和数值评估。这种方法的基础是只有在收集到足够的信息数据时才进行交流。

英文摘要:

Modern cyber-physical architectures use data collected from systems at different physical locations to learn appropriate behaviors and adapt to uncertain environments. However, an important challenge arises as communication exchanges at the edge of networked systems are costly due to limited resources. This paper considers a setup where multiple agents need to communicate efficiently in order to jointly solve a reinforcement learning problem over time-series data collected in a distributed manner. This is posed as learning an approximate value function over a communication network. An algorithm for achieving communication efficiency is proposed, supported with theoretical guarantees, practical implementations, and numerical evaluations. The approach is based on the idea of communicating only when sufficiently informative data is collected.

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。

↑