Off-Policy Actor-critic for Recommender Systems.
Minmin ChenCan XuVince GattoDevanshu JainAviral KumarEd H. ChiPublished in: RecSys (2022)
Keyphrases
- recommender systems
- actor critic
- reinforcement learning
- optimal control
- approximate dynamic programming
- policy gradient
- gradient method
- temporal difference
- collaborative filtering
- neuro fuzzy
- reinforcement learning algorithms
- function approximation
- policy iteration
- matrix factorization
- linear program
- dynamic programming
- evaluation function
- average reward
- belief revision