Unified Off-Policy Learning to Rank: a Reinforcement Learning Perspective.
Zeyu ZhangYi SuHui YuanYiran WuRishab BalasubramanianQingyun WuHuazheng WangMengdi WangPublished in: NeurIPS (2023)
Keyphrases
- learning to rank
- reinforcement learning
- ranking functions
- exploration exploitation dilemma
- information retrieval
- balancing exploration and exploitation
- loss function
- ranking svm
- document retrieval
- supervised learning
- direct optimization
- query dependent
- evaluation measures
- learning algorithm
- learning to rank algorithms
- machine learning
- collaborative filtering
- feature extraction
- evaluation metrics
- training set
- feature space
- test collection
- language model
- multi class