AI Paper Atlas

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

Convolution Monge Mapping Normalization for learning on sleep data

Theo Gnassounou, Rémi Flamary, Alexandre Gramfort

NeurIPS 2023 · Label-efficient and robust learning › Test-time adaptation · approach: Extrapolation and single-sample adaptation · 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
Use CMMN instead of Sample normalizationChambon · Cross-dataset meanBACC (%): 58.6 ± 2.6 -> 62.7 ± 2.4Improved
Use CMMN instead of Sample normalizationChambon · SHHS→MASSBACC (%): 57.0 ± 2.8 -> 64.5 ± 2.8Improved
Use CMMN instead of no normalizationChambon · Phys.→SHHSBACC (%): 53.9 ± 3.2 -> 51.6 ± 1.8Hurt
Use CMMN vs subject-specific normalizations (Conv, Norm, ConvNorm)Chambon · Physionet cross-sessionBACC (%): 68.1 ± 1.3 -> 74.8 ± 0.6Improved
Use subject-specific normalization instead of No AdaptChambon · Physionet cross-sessionBACC (%): 73.7 ± 0.7 -> 67.5 ± 2.7Hurt
Use CMMN with DeepSleepNetDeepSleepNet · Mean (9 pairs)BACC (%): 54.6 ± 2.9 -> 58.6 ± 3.2Improved
Use CMMN with DeepSleepNetDeepSleepNet · MASS→MASSBACC (%): 73.3 ± 1.7 -> 73.1 ± 2.6No clear effect
Add DANN on top of CMMNCMMN · Cross-dataset mean (6 pairs)BACC (%): 60.0 ± 2.7 -> 59.6 ± 1.7No clear effect
Add DANN on top of CMMNCMMN · Cross-dataset mean (6 pairs)ΔBACC@20 (%): 8.5 ± 4.4 -> 9.1 ± 4.9Improved
Add DANN alone to No AdaptChambon · Cross-dataset mean (6 pairs)BACC (%): 54.9 ± 4.5 -> 55.4 ± 1.6No clear effect
Map to power-law PSD instead of barycenter (ablation)CMMN · Mean (9 pairs), PSD targetBACC (%): 63.8 ± 2.9 -> 61.9 ± 6.4Hurt
Map to white-noise PSD (whitening) instead of barycenter (ablation)CMMN · Mean (9 pairs), PSD targetBACC (%): 63.8 ± 2.9 -> 61.9 ± 5.1Hurt

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