Policy Iteration-Based Conditional Termination and Ranking Functions.
Damien MasséPublished in: VMCAI (2014)
Keyphrases
- ranking functions
- policy iteration
- markov decision processes
- learning to rank
- model free
- optimal policy
- reinforcement learning
- fixed point
- document retrieval
- least squares
- sample path
- finite state
- web search engines
- supervised learning
- web search
- ranking algorithm
- temporal difference
- markov decision process
- infinite horizon
- linear programming
- optimal control
- convergence rate
- machine learning
- state space
- monte carlo
- feature set
- semi supervised
- image retrieval
- training data
- learning algorithm
- data sets