High Probability Guarantees for Nonconvex Stochastic Gradient Descent with Heavy Tails.

Shaojie Li Yong Liu

Published in: ICML (2022)

Keyphrases

stochastic gradient descent
heavy tails
loss function
step size
least squares
matrix factorization
probability distribution
support vector machine
image compression
objective function
collaborative filtering
unsupervised learning
weight vector
online algorithms