Implicit regularization in Heavy-ball momentum accelerated stochastic gradient descent.
Avrajit GhoshHe LyuXitong ZhangRongrong WangPublished in: ICLR (2023)
Keyphrases
- stochastic gradient descent
- least squares
- step size
- loss function
- matrix factorization
- regularization parameter
- random forests
- learning rate
- early stopping
- weight vector
- support vector machine
- multiple kernel learning
- online algorithms
- importance sampling
- convergence rate
- machine learning
- logistic regression
- missing data
- pairwise
- support vector
- feature extraction
- e learning