AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Contextual Reliability: When Different Features Matter in Different Contexts

Gaurav Rohit Ghosal, Amrith Setlur, Daniel S. Brown, Anca Dragan, Aditi Raghunathan

ICML 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 0 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
Standard group DRO (group definitions without context) instead of ERMCorrupted Waterbirds · Corrupted Waterbirdsworst-group accuracy: 0.67 to 0.60Hurt
Add ENP feature annotations on only 10% of training dataCorrupted Waterbirds · Corrupted Waterbirdsworst-group accuracy: 0.728 vs 0.73 for conDRO and GT-Aug. with full context labelsNo clear effect
Add ENP (feature-annotation-based two-stage) over ERMCorrupted Waterbirds · Corrupted Waterbirdsworst-group accuracy: 0.67 to 0.728Improved
Add ENP feature-based augmentation of the previous action on noisy MountainCarNoisy MountainCar · Noisy MountainCartest reward: -139.5 vs best baseline -170.4Improved
Provide previous action as input to the imitation policy (no augmentation)Noisy MountainCar · Noisy MountainCartest reward: -170.4 to -194Hurt
Add conDRO with ground-truth context on all pointsNoisy MountainCar · Noisy MountainCartest reward: -188.3 vs -194 with previous actionNo clear effect
Add annotated-spurious-agent augmentation (20% annotated) to MultiPath++WOMD MultiPath++ · Waymo Open Motion DatasetminADE (perturbed, lower is better): 0.817 to 0.801Improved
Random agent-deletion augmentation without annotationsWOMD MultiPath++ · Waymo Open Motion DatasetminADE (perturbed, lower is better): 0.817 to 0.815No clear effect
ENP: predicted spuriousness labels from a small feature predictor used to augment all trajectoriesWOMD MultiPath++ · Waymo Open Motion DatasetminADE (perturbed, lower is better): 0.801 to 0.774Improved
Increase fraction of non-spurious feature annotations in the simplified linear/deep setup (ablation)ENP on simplified two-context setup · Simplified two-context setuptest accuracyImproved
Use ENP feature-based context structure instead of ERM/IRM/conDRO on simplified setup (asymptotic)ENP vs conDRO, IRM, ERM · Simplified two-context setup (theory)asymptotic test accuracyImproved

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