AI Paper Atlas

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

Vibration-Based Uncertainty Estimation for Learning from Limited Supervision

Hengtong Hu, Lingxi Xie, Xinyue Huo, Richang Hong, Qi Tian

ECCV 2022 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Single-model predictive uncertainty · 2 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 label flipping sequence to the probability vibration measure (ablation)vibration uncertainty · Active learning CIFAR10 iter 1test error (%): 35.37 -> 27.33Improved
add label flipping sequence to the probability vibration measure (ablation)vibration uncertainty · Active learning CIFAR100 iter 1test error (%): 45.64 -> 43.99Improved
add class weights for pseudo label selection (ablation)vibration pseudo label mining with Mean Teacher · Mini-ImageNet SSL (4000 and 10000 labels)accuracy: 0.86% and 0.85% gainsImproved
add vibration-based pseudo labelsMean Teacher (RA) · CIFAR100 SSL (4000 labels)test error (%): 43.86 -> 36.86Improved
add vibration-based pseudo labelsFixMatch (RA) · CIFAR10 SSL (250 labels)test error (%): 6.16 -> 6.06Improved
add vibration-based pseudo labelsFixMatch (RA) · CIFAR100 SSL (4000 labels)test error (%): 31.22 -> 30.40Improved
replace MC Dropout uncertainty with vibration fusedactive learning on Mean Teacher · CIFAR10 iter 1test error (%): 33.27 -> 27.33Improved
replace K-center selection with vibration fusedactive learning on Mean Teacher · CIFAR10 iter 1test error (%): 35.53 -> 27.33Improved
start the sampled sequence at a later epoch instead of early epochs (ablation)vibration uncertainty · Active learning CIFAR10 and CIFAR100 iter 1test errorImproved
increase balance coefficient mu above 0.1 (ablation)vibration uncertainty · Active learning CIFAR10 and CIFAR100 iter 1test errorHurt
add vibration-based noisy sample handlingMean Teacher · CIFAR10 with 0.2 label noisetest error (%): 13.82 -> 10.06Improved
add vibration-based noisy sample handlingMean Teacher · CIFAR10 with 0.4 label noisetest error (%): 18.19 -> 12.77Improved

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