Accelerating Yade's poromechanical coupling with matrix factorization reuse, parallel task management, and GPU computing.
Robert A. CaulkEmanuele CatalanoBruno ChareyrePublished in: Comput. Phys. Commun. (2020)
Keyphrases
- matrix factorization
- collaborative filtering
- parallel implementation
- recommender systems
- low rank
- parallel computation
- parallel processing
- parallel computing
- missing data
- graphics processing units
- negative matrix factorization
- nonnegative matrix factorization
- factor analysis
- factorization methods
- parallel programming
- data sparsity
- probabilistic matrix factorization
- implicit feedback
- data matrix
- shared memory
- computer architecture
- tensor factorization
- data representation
- loss function
- latent factors
- variational bayesian
- item recommendation