AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Stein Self-Repulsive Dynamics: Benefits From Past Samples

Mao Ye, Tongzheng Ren, Qiang Liu

NeurIPS 2020 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Posterior sampling with diffusion guidance · 5 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
add Stein repulsive gradient to Langevin dynamicsLangevin dynamics · UCI Bostontest RMSE: 3.342 -> 3.086Improved
add Stein repulsive gradient to Langevin dynamicsLangevin dynamics · UCI Yachttest log-likelihood: -1.187 -> -0.458Improved
add Stein repulsive gradient to Langevin dynamicsLangevin dynamics · UCI WineWhitetest RMSE: 0.692 -> 0.688No clear effect
add repulsive gradient to Langevin dynamicsLangevin dynamics · UCI Navaltest log-likelihood: 4.041 -> 4.186Improved
add repulsive gradient to Langevin dynamicsLangevin dynamics · Mushroom banditcumulative regret: 4.28 -> 3.80Improved
add repulsive gradient to Langevin dynamicsLangevin dynamics · Wheel banditcumulative regret: 38.07 -> 32.08Improved
use SRLD instead of SVGDSVGD · Wheel banditcumulative regret: 91.32 -> 32.08Improved
use SRLD instead of SVGDSVGD · UCI Concretetest log-likelihood: -5.500 -> -3.034Improved
use SRLD instead of SVGDSVGD · UCI WineWhitetest RMSE: 0.655 -> 0.688Hurt
increase SRLD samples used from 20 to 100 (ablation)SRLD · UCI Bostontest RMSE: 3.236 -> 3.086Improved
increase SRLD samples used from 20 to 100 (ablation)SRLD · UCI Yachttest log-likelihood: -1.160 -> -0.458Improved
increase SRLD samples used from 20 to 100 (ablation)SRLD · UCI WineWhitetest RMSE: 0.695 -> 0.688No clear effect

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