Multi-Task Fusion via Reinforcement Learning for Long-Term User Satisfaction in Recommender Systems.
Qihua ZhangJunning LiuYuzhuo DaiYiyan QiYifan YuanKunlun ZhengFan HuangXianfeng TanPublished in: CoRR (2022)
Keyphrases
- user satisfaction
- multi task
- recommender systems
- long term
- reinforcement learning
- transfer learning
- learning problems
- collaborative filtering
- multi task learning
- information systems
- learning tasks
- multitask learning
- multiple tasks
- gaussian processes
- multi class
- feature selection
- matrix factorization
- sparse learning
- knowledge transfer
- machine learning
- perceived usefulness
- service quality
- learning process
- learning algorithm
- user perceptions
- user preferences
- semi supervised learning
- supervised learning
- social networks
- information quality
- data mining
- kolmogorov smirnov