UMA-MF: A Unified Multi-CPU/GPU Asynchronous Computing Framework for SGD-Based Matrix Factorization.
Yizhi HuangYan LiuYang BaiSi ChenRenfa LiPublished in: IEEE Trans. Parallel Distributed Syst. (2023)
Keyphrases
- matrix factorization
- collaborative filtering
- stochastic gradient descent
- low rank
- recommender systems
- missing data
- data sparsity
- negative matrix factorization
- nonnegative matrix factorization
- item recommendation
- data matrix
- factor analysis
- implicit feedback
- tensor factorization
- latent factor models
- variational bayesian
- parallel computing
- latent factors
- graphics processors
- factorization methods
- pairwise