典型文献
Improving deep reinforcement learning by safety guarding model via hazardous experience planning
文献摘要：
1 Introduction and main contributions
Deep reinforcement learning that considers the advantages of both deep learning and reinforcement learning has achieved success in many fields[l].However,during the learning process,a possibility still exists that the agent fails in the task because of falling into hazardous states due to taking improper actions.It can be concluded that setting up a mechanism to avoid these hazardous states and actions is in a position to improve the success rate of the agent[2-4].How to ensure the safety of policies is a fuindamental obstacle to the practical application of artificial intelligence[5].The experience sam-ples are obtained through trial and error without an early warning model,and agents may get to local hazardous states[6],making it necessary to construct an effective safety guarding model that provides danger warning information and improve the robustness of the system.
文献关键词：
中图分类号：
作者姓名：
Pai PENG;Fei ZHU;Xinghong LING;Peiyao ZHAO;Quan LIU
作者机构：
School of Computer Science and Technology,Soochow University,Suzhou 215006,China
文献出处：
引用格式：
[1]Pai PENG;Fei ZHU;Xinghong LING;Peiyao ZHAO;Quan LIU-.Improving deep reinforcement learning by safety guarding model via hazardous experience planning)[J].计算机科学前沿,2022(04):214-216
A类：
Deep,fuindamental
B类：
Improving,deep,reinforcement,learning,by,safety,guarding,model,via,hazardous,experience,planning,Introduction,main,contributions,that,considers,advantages,both,has,achieved,success,many,fields,However,during,process,possibility,still,exists,fails,task,because,falling,into,states,due,taking,improper,actions,It,can,concluded,setting,up,mechanism,avoid,these,position,improve,rate,ensure,policies,obstacle,practical,application,artificial,intelligence,sam,ples,are,obtained,through,trial,error,without,early,warning,agents,may,get,local,making,necessary,construct,effective,provides,danger,information,robustness,system
AB值：
0.607719
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
