Stochastic Gradient Descent on Highly-Parallel Architectures.
Yujing MaFlorin RusuMartin TorresPublished in: CoRR (2018)
Keyphrases
- highly parallel
- stochastic gradient descent
- parallel architectures
- matrix factorization
- least squares
- step size
- loss function
- efficient implementation
- parallel processing
- random forests
- computing systems
- single pass
- regularization parameter
- single chip
- support vector machine
- multiple kernel learning
- weight vector
- online algorithms
- parallel programming
- massively parallel
- general purpose
- parallel computers
- importance sampling
- collaborative filtering
- data sets
- image restoration
- computer systems
- linear svm
- super resolution
- low cost
- bayesian networks
- decision trees
- machine learning