Continuous improvement of a document treatment chain using reinforcement learning.
Esther NicartBruno ZanuttiniBruno GrilhèresPatrick GirouxPublished in: IC (2015)
Keyphrases
- reinforcement learning
- action space
- information retrieval
- information retrieval systems
- function approximation
- document collections
- document images
- retrieval systems
- state space
- document retrieval
- learning process
- supervised learning
- model free
- document clustering
- optimal control
- vector space model
- fitted q iteration
- continuous state spaces
- continuous state and action spaces
- database
- transfer learning
- significant improvement
- machine learning
- user queries
- web documents
- query expansion
- co occurrence
- continuous domains
- document analysis
- document content
- multi agent
- learning algorithm