AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Online Adaptation to Label Distribution Shift

Ruihan Wu, Chuan Guo, Yi Su, Kilian Q Weinberger

NeurIPS 2021 · Label-efficient and robust learning › Test-time adaptation · approach: Robust optimization for distribution shift · 8 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 online gradient descent for reweighting instead of the fixed classifierResNet-18 · CIFAR-10 simulated label shiftsaverage error (%)Improved
Use OGD with surrogate-loss gradients instead of finite-difference gradients (ablation)OGD · CIFAR-10 simulated label shiftsaverage error (%): 10.71 ± 0.10 -> 10.24 ± 0.06Improved
Use OGD with finite-difference gradients rather than surrogate-loss gradients (ablation)OGD · ArXiv test setaverage error (%): 25.70 -> 25.52Improved
Use FTH instead of the optimal fixed classifierLabel-shift adaptation · CIFAR-10 simulated label shiftsaverage error (%): 10.24 ± 0.08 -> 10.25 ± 0.06No clear effect
Use OGD instead of the optimal fixed classifierLabel-shift adaptation · CIFAR-10 monotone shiftaverage error (%): 10.24 ± 0.08 -> 9.75 ± 0.07Improved
Use FTFWH with window size 100 when the periodicity is 100 (ablation)FTFWH · CIFAR-10 periodic shift (Tp=100)average error (%): 10.25 ± 0.08 -> 12.15 ± 0.01Hurt
Change FTFWH window size from 10^2 to 10^3 (ablation)FTFWH · ArXiv test setaverage error (%): 30.14 -> 26.09Improved
Use FTFWH instead of FTH or OGDOnline label-shift adaptation · ArXiv test setaverage error (%): 25.62 -> 30.14Hurt
Use OGD instead of the optimal fixed classifierLabel-shift adaptation · ArXiv test setaverage error (%): 25.56 -> 25.52No 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