Investigating Action-Space Generalization in Reinforcement Learning for Recommendation Systems.
Abhishek NaikBo ChangAlexandros KaratzoglouMartin MladenovEd H. ChiMinmin ChenPublished in: WWW (Companion Volume) (2023)
Keyphrases
- action space
- recommendation systems
- reinforcement learning
- state space
- markov decision processes
- state and action spaces
- real valued
- collaborative filtering
- continuous state
- user preferences
- reinforcement learning methods
- stochastic processes
- state action
- recommender systems
- function approximation
- action selection
- web search
- collaborative filtering recommendation algorithm
- user feedback
- continuous state spaces
- search engine
- function approximators
- markov decision process
- reinforcement learning algorithms
- multi agent
- information retrieval
- model free
- learning algorithm
- dynamic programming
- optimal policy
- learning problems
- control problems
- markov decision problems
- single agent
- partially observable