电子学报 ›› 2021, Vol. 49 ›› Issue (2): 372-379.DOI: 10.12263/DZXB.20200139

• 综述评论 • 上一篇    下一篇

基于神经进化的深度学习模型研究综述

韩冲, 王俊丽, 吴雨茜, 张超波   

  1. 同济大学电子与信息工程学院, 上海 201804
  • 收稿日期:2020-04-17 修回日期:2020-05-10 出版日期:2021-02-25 发布日期:2021-02-25
  • 通讯作者: 王俊丽
  • 作者简介:韩冲 女,1996年11月出生于山西临汾.2018年进入同济大学电子与信息工程学院,在读硕士研究生.研究方向为深度学习、进化计算.E-mail:496274966@qq.com
  • 基金资助:
    国家重点研发计划(No.2017YFA0700602);国家自然科学基金(No.61672381);中央高校基本科研基金

A Review of Deep Learning Models Based on Neuroevolution

HAN Chong, WANG Jun-li, WU Yu-xi, ZHANG Chao-bo   

  1. College of Electronics and Information Engineering, University of Tongji, Shanghai 201804, China
  • Received:2020-04-17 Revised:2020-05-10 Online:2021-02-25 Published:2021-02-25

摘要: 深度学习研究发展至今已可以胜任各类识别、分类、生成任务,但是对于不同的任务,神经网络的结构或参数不可能只是微小的变化,依然需要专家进行调整.在这样的情况下,自动化地调整神经网络的结构或参数成为研究热点.其中,以达尔文自然进化论为灵感的神经进化成为主要优化方法.利用神经进化优化的深度学习模型以种群为基础,通过突变、重组等操作进化,可实现自动地、逐步地构建神经网络并最终选择出性能最优的深度学习模型.本文简述了神经进化与进化计算;详细概述了各类基于神经进化的深度学习模型;分析了各类模型的性能;总结了神经进化与深度学习融合的前景并探讨下一步的研究方向.

 

关键词: 神经进化, 深度学习, 进化计算, 卷积神经网络, 生成式对抗网络, 自动编码器, 长短期记忆网络, 深度强化学习

Abstract: With the development of deep learning,it has been competent to various types of tasks,such as identification,classification,and generation.However,the structures or parameters of artificial neural networks cannot be only a little changed for different task,experts need to adjust the structures or parameters of the neural network.In such situations,the method of automatically adjusting the structures or parameters of the artificial neural network has become a research hotspot,among these methods,neuroevolution inspired by Darwin's natural evolution theory has become the main optimization method for that.Deep learning models optimized by neuroevolution based on population,evolving through mutation,crossover and other operations,can automatically and gradually construct the neural network and then choose the most optimal deep learning model.This paper summarizes the neuroevolution and the evolutionary computation.It elaborates various deep learning models based on neuroevolution,and analyzes the performance of these models.It concludes prospects of the deep learning model based on neuroevolution and discusses the next research directions.

Key words: neuroevolution, deep learning, evolutionary computation, convolutional neural networks, generative adversarial networks, autoencoder, long short-term memory networks, deep reinforcement learning

中图分类号: