AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Auxiliary Task Reweighting for Minimum-data Learning

Baifeng Shi, Judy Hoffman, Kate Saenko, Trevor Darrell, Huijuan Xu

NeurIPS 2020 · Label-efficient and robust learning › Semi-supervised learning · approach: Self-training theory · 14 citations (OpenAlex)

Read the paper

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
Reweight Rotation and Exemplar-MT auxiliary tasks with ARML instead of uniform weightsS4L · CIFAR-10 (4000 labels)test error: 15.67 ± .29 -> 13.68 ± .35Improved
Reweight Rotation and Exemplar-MT auxiliary tasks with ARML instead of uniform weightsS4L · SVHN (1000 labels)test error: 7.83 ± .33 -> 5.89 ± .22Improved
Use ARML to reweight auxiliary tasks instead of the baseline weightingCelebA 5_o_Clock_Shadow · CelebA (1% main-task labels)test error: 6.70 ± .18 -> 5.97 ± .18Improved
Use ARML to reweight source domains without adding target labels to the classification lossPACS domain generalization Baseline · PACS Sketch target (5 extra labels per class)accuracy: 75.34 -> 79.35Improved
Add ARML reweighting of source domainsPACS joint training · PACS Art target (5 extra labels per class)accuracy: 83.94 -> 85.70Improved
Add ARML reweighting of source domainsPACS joint training · PACS Cartoon target (5 extra labels per class)accuracy: 81.36 -> 81.01Hurt
Add ARML reweighting of source domainsPACS joint training · PACS Photo target (5 extra labels per class)accuracy: 97.01 -> 97.22Improved
Add ARML reweighting of source domainsPACS joint training · PACS Sketch target (5 extra labels per class)accuracy: 78.52 -> 80.47Improved

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