AI Paper Atlas

Research map / Optimization / Optimizers

Statistical Inference Using SGD

Tianyang Li, Liu Liu, Anastasios Kyrillidis, Constantine Caramanis

AAAI 2018 · Optimization › Optimizers · approach: Stochastic gradient MCMC · 27 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 η=0.1 rather than η=0.004 with t=2500 (ablation)SGD inference · Linear regression synthetic, Experiment 1coverage probability: 0.916 -> 0.960Improved
Use η=0.1 rather than η=0.004 with t=2500 (ablation)SGD inference · Linear regression synthetic, Experiment 1average confidence interval width: 3.70 -> 4.53Improved
Use t=2500 rather than t=100 at η=0.1 (ablation)SGD inference · Linear regression synthetic, Experiment 1coverage probability: 0.957 -> 0.960No clear effect
Use η=0.1 rather than η=0.004 with t=2500 (ablation)SGD inference · Linear regression synthetic, Experiment 2coverage probability: 0.897 -> 0.963Improved
Use η=0.1 rather than η=0.004 with t=2500 (ablation)SGD inference · Logistic regression synthetic, Experiment 1coverage probability: 0.86 -> 0.939Improved
Use η=0.1 rather than η=0.004 with t=2500 (ablation)SGD inference · Logistic regression synthetic, Experiment 2coverage probability: 0.851 -> 0.947Improved
Use η=0.1 rather than η=0.004 with t=2500 (ablation)SGD inference · Logistic regression synthetic, Experiment 1average confidence interval width: 0.194 -> 0.258Mixed
Use η=0.1 rather than η=0.004 with t=2500 (ablation)SGD inference · Logistic regression synthetic, Experiment 2average confidence interval width: 0.195 -> 0.266Mixed
Use SGD inference instead of bootstrapSGD inference · Linear regression synthetic, Experiment 1coverage probability: 0.941 -> 0.960Improved
Use SGD inference instead of bootstrapSGD inference · Linear regression synthetic, Experiment 2coverage probability: 0.938 -> 0.963Improved
Use SGD inference instead of bootstrapSGD inference · Logistic regression synthetic, Experiment 1coverage probability: 0.932 -> 0.939Improved
Use SGD inference instead of normal approximationSGD inference · Logistic regression synthetic, Experiment 1coverage probability: 0.957 -> 0.939Hurt

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