Edge-Compatible Reinforcement Learning for Recommendations.
James E. KostasPhilip S. ThomasGeorgios TheocharousPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- recommender systems
- function approximation
- edge information
- state space
- reinforcement learning algorithms
- learning algorithm
- edge detection
- optimal policy
- markov decision processes
- model free
- edge detector
- machine learning
- markov decision process
- edge direction
- disjoint paths
- temporal difference
- image processing
- learning problems
- transfer learning
- dynamic programming
- image gradient
- partially observable
- multi agent
- multi agent reinforcement learning
- case study