Non-stationary Dueling Bandits for Online Learning to Rank.
Shiyin LuYuan MiaoPing YangYao HuLijun ZhangPublished in: APWeb/WAIM (2) (2022)
Keyphrases
- non stationary
- learning to rank
- balancing exploration and exploitation
- ranking functions
- information retrieval
- loss function
- evaluation measures
- ranking svm
- online learning
- adaptive algorithms
- direct optimization
- evaluation metrics
- document retrieval
- collaborative filtering
- empirical mode decomposition
- retrieval systems
- test collection
- query dependent
- similarity measure
- reinforcement learning
- directly optimize
- supervised learning
- concept drift
- information retrieval systems
- change point detection
- learning to rank algorithms
- web pages