Tuning Synaptic Connections Instead of Weights by Genetic Algorithm in Spiking Policy Network  

作　　者：Duzhen Zhang Tielin Zhang Shuncheng Jia Qingyu Wang Bo Xu 

机构地区：[1]School of Artificial Intelligence,University of Chinese Academy of Sciences,Beijing,100049,China [2]Institute of Automation,Chinese Academy of Sciences(CAS),Beijing,100190,China [3]Center for Excellence in Brain Science and Intelligence Technology,Chinese Academy of Sciences(CAS),Shanghai,200031,China

出　　处：《Machine Intelligence Research》2024年第5期906-918,共13页机器智能研究（英文版）

基　　金：supported by the Beijing Nova Program,China(No.20230484369);the Strategic Priority Research Program of Chinese Academy of Sciences,China(No.XDA27010404);the Shanghai Municipal Science and Technology Major Project,China(No.2021SHZDZX),the Youth Innovation Promotion Association of the Chinese Academy of Sciences,China.

摘　　要：Learning from interaction is the primary way that biological agents acquire knowledge about their environment and themselves.Modern deep reinforcement learning(DRL)explores a computational approach to learning from interaction and has made significant progress in solving various tasks.However,despite its power,DRL still falls short of biological agents in terms of energy efficiency.Although the underlying mechanisms are not fully understood,we believe that the integration of spiking communication between neurons and biologically-plausible synaptic plasticity plays a prominent role in achieving greater energy efficiency.Following this biological intuition,we optimized a spiking policy network(SPN)using a genetic algorithm as an energy-efficient alternative to DRL.Our SPN mimics the sensorimotor neuron pathway of insects and communicates through event-based spikes.Inspired by biological research showing that the brain forms memories by creating new synaptic connections and rewiring these connections based on new experiences,we tuned the synaptic connections instead of weights in the SPN to solve given tasks.Experimental results on several robotic control tasks demonstrate that our method can achieve the same level of performance as mainstream DRL methods while exhibiting significantly higher energy efficiency.

关 键 词：Spiking neural networks genetic evolution bio-inspired learning agent&cognitive architectures robotic control 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]