Toward Simulating Environments in Reinforcement Learning Based Recommendations.
Xiangyu ZhaoLong XiaZhuoye DingDawei YinJiliang TangPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- function approximation
- dynamic environments
- real world
- reinforcement learning algorithms
- highly dynamic
- model free
- learning process
- recommender systems
- markov decision processes
- web services
- learning algorithm
- real time
- multi agent environments
- state space
- optimal control
- temporal difference
- machine learning
- learning capabilities