AI Paper Atlas

Research map / Optimization / Optimizers

Decentralized Langevin Dynamics for Bayesian Learning

Anjaly Parayil, He Bai, Jemin George, Prudhvi Gurram

NeurIPS 2020 · Optimization › Optimizers · approach: Stochastic gradient MCMC · 10 citations (OpenAlex)

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
Use D-ULA with 10 agents instead of centralized ULABayesian logistic regression · a9atest accuracy (%): 83.89 -> 84.5637Improved
Use D-ULA instead of centralized ULABayesian logistic regression · a9aconvergence speedImproved
Increase number of agents from 1 to 10 (ablation)D-ULA · Gaussian mixtureSinkhorn distance: 0.259 -> 0.244Improved
Use D-ULA (5 agents) instead of SGDLeNet-5 · MNISTtest accuracy (%): 98.15 -> 98.52Improved
Use Langevin sampling (C-ULA/D-ULA) instead of SGD point estimateLeNet-5 · SVHN OOD detectionmean predicted probability (lower better on OOD): 0.849 -> 0.651Improved
Use D-ULA instead of C-ULA for OOD detectionLeNet-5 · SVHN OOD detectionmean predicted probability: 0.849 -> 0.604Improved

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