Divide and Conquer: Towards Better Embedding-based Retrieval for Recommender Systems From a Multi-task Perspective.
Yuan ZhangXue DongWeijie DingBiao LiPeng JiangKun GaiPublished in: CoRR (2023)
Keyphrases
- multi task
- recommender systems
- multi task learning
- learning tasks
- collaborative filtering
- multiple tasks
- feature selection
- transfer learning
- multi class
- multitask learning
- sparse learning
- metric learning
- information retrieval
- gaussian processes
- learning problems
- image retrieval
- matrix factorization
- kernel methods
- information gain
- data sets
- theoretical analysis
- support vector
- machine learning
- data mining