深度强化学习综述

王浩楠, 刘苧, 章艺云, 冯大伟, 黄峰… - 信息与电子工程前沿 …, 2022 - fitee.zjujournals.com
gradient method has the problem that the policy is difficult to update stably in the case of
unstable data because of using a neural network … multiple shunt networks to randomize different …

[PDF][PDF] 基于非线性时间序列分析的被动动态行走混沌现象研究

WUN TIME-SERIES - 2016 - researchgate.net
… Chaotic Optimization Algorithm (COA): Scientists took a great interest in developing
optimization techniques based on chaotic search routines, eg, chaotic neural network, chaotic …