A novel movies recommendation algorithm based on reinforcement learning with DDPG policy.
Qiaoling ZhouPublished in: Int. J. Intell. Comput. Cybern. (2020)
Keyphrases
- reinforcement learning
- recommendation algorithms
- optimal policy
- policy search
- filtering algorithm
- action selection
- markov decision process
- partially observable environments
- recommendation systems
- recommender systems
- collaborative filtering
- reinforcement learning algorithms
- actor critic
- reward function
- markov decision problems
- reinforcement learning problems
- markov decision processes
- function approximation
- policy gradient
- user preferences
- partially observable
- state space
- state and action spaces
- control policy
- social recommendation
- action space
- policy evaluation
- partially observable markov decision processes
- inverse reinforcement learning
- function approximators
- average reward
- optimal control
- dynamic programming
- policy iteration
- temporal difference
- agent learns
- infinite horizon
- collaborative filtering recommendation
- state action
- personalized recommendation
- cold start
- learning algorithm