On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,051 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “SVAG”

Cocoercivity, Smoothness and Bias in Variance-Reduced Stochastic Gradient Methodsmath.OC

SVAG method examines biased updates in variance-reduced stochastic gradient methods.

problem Examining biased updates in variance-reduced stochastic gradient methods.

method Introduces SVAG, a SAG/SAGA-like method with adjustable bias, analyzed in a cocoercive root-finding setting.

result SVAG's step-size requirements for gradients are less restrictive compared to non-gradient cases, highlighting the need for careful analysis.

SVAG biased updates variance-reduced stochastic gradient cocoercivity

2019-03-21abs ↗pdf ↗

On the Validity of Modeling SGD with Stochastic Differential Equations (SDEs)cs.LG

The paper clarifies the approximation of SGD with Ito SDEs for finite learning rates.

problem Theoretical justification and experimental verification of the Ito SDE approximation for finite learning rates in SGD.

method An efficient simulation algorithm SVAG and a necessary condition test for the SDE approximation.

result The Ito SDE approximation can meaningfully capture training and generalization properties of deep nets with finite learning rates.

SGD Stochastic Differential Equations Finite Learning Rate Generalization Deep Nets

2021-02-24abs ↗pdf ↗