RLMixer: A Reinforcement Learning Approach for Integrated Ranking with Contrastive User Preference Modeling.
Jing WangMengchen ZhaoWei XiaZhenhua DongRuiming TangRui ZhangJianye HaoGuangyong ChenPheng-Ann HengPublished in: PAKDD (3) (2023)
Keyphrases
- user preferences
- reinforcement learning
- user feedback
- user specific
- preference model
- web search
- user profiles
- making recommendations
- collaborative filtering
- user behavior
- recommender systems
- ranking algorithm
- recommendation systems
- social influence
- user behaviour
- preference learning
- learning algorithm
- markov decision processes
- optimal policy
- model free
- preference models