Convergence diagnostics for stochastic gradient descent with constant learning rate.
Jerry CheePanos ToulisPublished in: AISTATS (2018)
Keyphrases
- learning rate
- stochastic gradient descent
- convergence rate
- step size
- weight vector
- update rule
- convergence speed
- convergence theorem
- rapid convergence
- training speed
- adaptive learning rate
- learning algorithm
- least squares
- matrix factorization
- loss function
- perceptron algorithm
- random forests
- training data
- online algorithms
- regularization parameter
- particle swarm optimization