A Nonlinear Time-Series Theory for Constant-step SGD with Momentum

stat.ML, cs.LG

Submitted: 2026-10-12

Updated: 2026-10-12

Comments: 42 pages, 2 figures

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• Acceleration of stochastic gradient descent with momentum by averaging: finite-sample rates and asymptotic normality

Related papers

• Behavior of prediction performance metrics with rare events
• Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
• A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
• Efficient quadratic entropy with distance sketches
• Embedding-Bias in Conditional Independence Testing
• From Log-Odds to Shapley Values: An Explanatory Geometry for the Weighted Naive Bayes Classifier