Global Convergence and Stability of Stochastic Gradient Descent.
Vivak PatelShushu ZhangBowen TianPublished in: NeurIPS (2022)
Keyphrases
- global convergence
- stochastic gradient descent
- step size
- convergence rate
- convergence speed
- global optimum
- optimization methods
- loss function
- least squares
- matrix factorization
- random forests
- importance sampling
- support vector machine
- differential evolution
- learning rate
- online algorithms
- weight vector
- particle swarm
- regularization parameter
- pso algorithm
- particle swarm optimization
- multiple kernel learning
- monte carlo
- support vector
- pairwise
- negative matrix factorization
- neural network
- hybrid algorithm
- decision trees
- feature selection