AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Tractable Approximate Gaussian Inference for Bayesian Neural Networks

James-A. Goulet, Luong Ha Nguyen, Saeid Amiri

JMLR 2021 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Infinite-width networks as Gaussian processes · 17 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 full instead of diagonal hidden-unit covariance (ablation)TAGI, 1D toy · 1D toy regression y=x^3+vpropagation of input uncertainty: diagonal covariances unable to retrieve reference std 6x^2/10; full covariance follows it closelyImproved
increase hidden units per layer from 5 to 50 (ablation)TAGI with full covariance · 1D toy regressionposterior correlation between parameters: non-zero correlations from extensive to a small fraction of the covariance matrixImproved
use diagonal instead of full covariance with many hidden units (>100) (ablation)TAGI · sum-of-units toy modelposterior correlation of hidden units: non-negligible for <10 units, almost zero for >100No clear effect
increase MNIST hidden units from 100 to 800 (B=1) (ablation)TAGI · MNIST testclassification error (%): 2.29 -> 1.54Improved
increase MNIST hidden units from 100 to 800 (B=10) (ablation)TAGI · MNIST testclassification error (%): 2.45 -> 1.53Improved
train to early-stopped epoch instead of one epoch (A=800, B=1) (ablation)TAGI · MNIST testclassification error (%): 3.15 -> 1.54Improved
train to early-stopped epoch instead of one epoch (A=100, B=10) (ablation)TAGI · MNIST testclassification error (%): 3.56 -> 2.45Improved
use batch size 10 instead of 1 (A=800) (ablation)TAGI · MNIST testclassification error (%): 1.54 -> 1.53No clear effect
use batch size 10 instead of 1 (A=100) (ablation)TAGI · MNIST testclassification error (%): 2.29 -> 2.45Hurt
compare TAGI with VMG on EnergyTAGI · Energy (UCI regression)RMSE: 1.46 vs 0.54 (VMG)Hurt
compare TAGI with VMG on ConcreteTAGI · Concrete (UCI regression)RMSE: 5.72 vs 4.89 (VMG)Hurt
compare TAGI with MC-Dropout on BostonTAGI · Boston (UCI regression)RMSE: 2.98 vs 2.97 (MC-Dropout)No 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