Variance Reduction in Gradient Exploration for Online Learning to Rank.
Huazheng WangSonwoo KimEric McCord-SnookQingyun WuHongning WangPublished in: SIGIR (2019)
Keyphrases
- learning to rank
- balancing exploration and exploitation
- variance reduction
- gradient estimation
- ranking functions
- reinforcement learning
- information retrieval
- loss function
- monte carlo
- evaluation measures
- document retrieval
- ranking svm
- sample size
- collaborative filtering
- learning to rank algorithms
- test collection
- web search engines
- importance sampling
- document collections
- precision and recall
- model selection
- search engine
- machine learning