High-probability Convergence Bounds for Nonlinear Stochastic Gradient Descent Under Heavy-tailed Noise.
Aleksandar ArmackiPranay SharmaGauri JoshiDragana BajovicDusan JakoveticSoummya KarPublished in: CoRR (2023)
Keyphrases
- heavy tailed
- stochastic gradient descent
- matrix factorization
- generalized gaussian
- least squares
- loss function
- probability distribution
- step size
- lower bound
- upper bound
- missing data
- noise level
- convergence rate
- importance sampling
- noise reduction
- image processing
- linear combination
- weight vector
- particle filter
- support vector machine
- bayesian networks