Generator and Critic: A Deep Reinforcement Learning Approach for Slate Re-ranking in E-commerce.
Jianxiong WeiAnxiang ZengYueqiu WuPeng GuoQingsong HuaQingpeng CaiPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- function approximation
- actor critic
- reinforcement learning algorithms
- temporal difference
- ranking algorithm
- electronic commerce
- web search
- model free
- policy gradient
- product search
- multi agent
- state space
- ranking functions
- learning to rank
- learning process
- evaluation function
- optimal control
- markov decision processes
- supervised learning
- reinforcement learning methods
- policy search
- information retrieval
- machine learning
- learning problems
- user feedback
- link analysis
- optimal policy
- temporal difference learning
- dynamic programming
- approximate dynamic programming