Recommendations with Negative Feedback via Pairwise Deep Reinforcement Learning.
Xiangyu ZhaoLiang ZhangZhuoye DingLong XiaJiliang TangDawei YinPublished in: KDD (2018)
Keyphrases
- pairwise
- negative feedback
- reinforcement learning
- positive feedback
- relevance feedback
- function approximation
- recommender systems
- negative samples
- state space
- pairwise interactions
- optimal policy
- similarity measure
- machine learning
- positive and negative feedback
- temporal difference
- markov decision processes
- multi agent
- learning algorithm
- transfer learning
- user preferences
- reinforcement learning algorithms
- reinforcement learning methods
- markov random field
- information retrieval
- data sets