Correlated Noise in Epoch-Based Stochastic Gradient Descent: Implications for Weight Variances.
Marcel KühnBernd RosenowPublished in: CoRR (2023)
Keyphrases
- correlated noise
- stochastic gradient descent
- weight vector
- least squares
- loss function
- matrix factorization
- step size
- random forests
- learning rate
- multiple kernel learning
- regularization parameter
- noise free
- online algorithms
- hyperplane
- importance sampling
- support vector machine
- linear combination
- cost function
- learning algorithm
- gaussian distribution