AI Paper Atlas

Research map / Optimization / Optimizers

Exploration of the (Non-)Asymptotic Bias and Variance of Stochastic Gradient Langevin Dynamics

Sebastian J. Vollmer, Konstantinos C. Zygalakis, Yee Whye Teh

JMLR 2016 · Optimization › Optimizers · approach: Stochastic gradient MCMC

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add a bias-correcting term to SGLD (mSGLD) removing first-order gradient-variance biasSGLD · Gaussian toy model, second momentasymptotic bias / MSE2Improved
use mSGLD instead of SGLD when the number of steps is large (high accuracy regime)SGLD · Gaussian toy model, optimal costcomputational cost for MSE2Improved
use mSGLD with small data batches (n = N^0.1, N^0.5)SGLD · Gaussian toy model, growing NMSE2Hurt
vary subset size for mSGLD in a Bayesian logistic-style example: n=150, n=50, n=10 (ablation)mSGLD · 2D example, posterior mean, h = 0.002MSEMixed
subsample the data (SGLD) instead of full gradients (Euler) as N growsEuler method · Gaussian toy model, N to infinitycomputational cost for vanishing MSE2Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy