Research map / Probabilistic and scientific modeling / Molecules and proteins
Evaluating Latent Space Robustness and Uncertainty of EEG-ML Models under Realistic Distribution Shifts
Neeraj Wagh, Jionghao Wei, Samarth Rawal, Brent M. Berry, Yogatheesan Varatharajah
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Evaluate on out-of-sample NMT data instead of TUAB | FSE · EEG grade | AUC: 0.920 -> 0.715 | Hurt |
| Evaluate on out-of-sample NMT data instead of TUAB | FSE · Brain age | MAE: 9.470 -> 12.411 | Hurt |
| Add Gaussian noise sigma=0.1 to input | PSDE · EEG grade | AUC: 0.767 -> 0.669 | Hurt |
| Add Gaussian noise sigma=0.001 to input | SSE/FSE/PSDE · EEG grade | AUC: 0.769 -> 0.769 | No clear effect |
| Quantization depth D=12, 8 or 6 | SSE/FSE/PSDE · EEG grade | AUC: 0.920 -> 0.920 | No clear effect |
| Band-pass filter f=[0.5, 30] | FSE · Brain age | MAE: 9.470 -> 11.239 | Hurt |
| Band-pass filter f=[1, 30] | SSE · EEG grade | AUC: 0.769 -> 0.778 | Improved |
| Baseline noise sigma=0.01 | SSE · EEG grade | AUC: 0.769 -> 0.784 | Improved |
| Baseline noise sigma=0.1 | FSE · Brain age | MAE: 9.470 -> 14.048 | Hurt |
| Baseline noise sigma=0.1 | PSDE · Brain age | MAE: 15.560 -> 14.662 | Improved |
| Baseline noise sigma=0.1 | all encoders · Latent integrity | integrity score: 0.53 -> 0.04 | Hurt |
| Band-pass filter f=[1, 25] | FSE-Grade · Latent integrity | integrity score: 0.49 -> 0.36 | Hurt |
Closest papers by meaning
- Geodesic Optimization for Predictive Shift Adaptation on EEG data (NeurIPS 2024 · similarity 0.86)
- NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces (NeurIPS 2026 · similarity 0.86)
- What do EEG Foundation Models Capture from Human Brain Signals? (NeurIPS 2026 · similarity 0.86)
- Is Limited Participant Diversity Impeding EEG-based Machine Learning? (NeurIPS 2025 · similarity 0.85)
- CLEF: EEG Foundation Models for Learning Clinical Semantics (NeurIPS 2026 · similarity 0.85)
- LUNA: Efficient and Topology-Agnostic Foundation Model for EEG Signal Analysis (NeurIPS 2025 · similarity 0.84)
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.
