AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

HEEGNet: Hyperbolic Embeddings for EEG

Shanglin Li, Chu Shiwen, Okan Koç, Yi Ding, Qibin Zhao, Motoaki Kawanabe, Ziheng Chen

ICLR 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Augmentation and mixture-of-experts generalization · 1 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
replace Euclidean MLR with hyperbolic MLR (ablation)EEGNet · Facedbalanced accuracy (%): 24.8 -> 38.8Improved
replace Euclidean MLR with hyperbolic MLR (ablation)EEGNet · Nakanishibalanced accuracy (%): 57.2 -> 60.8Improved
replace Euclidean MLR with hyperbolic MLR (ablation)EEGNet · Seedbalanced accuracy (%): 74.5 -> 75.1No clear effect
use HEEGNet with DSMDBN+EA instead of EEGNetEEGNet · overall cross-domain EEGbalanced accuracy (%): 39.9 -> 81.7Improved
add moment alignment in hyperbolic space (ablation)HEEGNet · overall cross-domain EEGbalanced accuracy (%): -12.3 to -15.5 withoutImproved
add distribution alignment (HHSW) on top of moment alignment (ablation)HEEGNet · overall cross-domain EEGbalanced accuracy (%): -5.0 withoutImproved
add distribution alignment without moment alignment (ablation)HEEGNet · overall cross-domain EEGbalanced accuracy (%): -12.3 vs -12.9No clear effect
add input-space EA to DSMDBN (ablation)HEEGNet · overall cross-domain EEGbalanced accuracy (%): 78.7 -> 81.7Improved
add EA input alignmentEEGNet · overall cross-domain EEGbalanced accuracy (%): 39.9 -> 67.4Improved
add STMA input alignmentFBCNet · overall cross-domain EEGbalanced accuracy (%): 31.6 -> 32.9No clear effect
add DSMDBN+EA vs DSMDBN alone (ablation)HEEGNet · VEPbalanced accuracy (%): 79.6 -> 77.4Hurt

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