Unassigned · 2026-11-03
Robust EEG-based Emotion Recognition from Coarse Trial-level Labels under Intra-trial Affective Variability
Xiaocong Zeng, Buxuan Wang, Xin Yang, Pei Huang, Dongyang Kuang
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
VARY is a variability-aware robust learning framework for EEG-based emotion recognition when only coarse trial-level labels are available. The authors use Two-stage Sample Selection (TSS) to select windows with reliable propagated labels, via a Label-Prediction Consistency Criterion (LPCC) and confidence filtering, and introduce an EEG-tailored time-series augmentation (MixAug). On label-noise learning on DREAMER and SEED, VARY reports improved test accuracy over best baselines.
Why it matters
The paper targets a practical EEG annotation mismatch: trial-level emotion labels are commonly propagated to short windows, even though emotions can fluctuate within a trial, creating window-level label noise. VARY aims to improve robustness under this setting by explicitly selecting reliable windows and applying EEG-specific augmentation, addressing challenges of low-SNR EEG signals and unreliable supervision.
Method
- Propose VARY for K-class EEG emotion classification under noisy labels from propagating coarse trial-level labels to short segments.
- Two-stage Sample Selection (TSS): Stage 1 uses Label-Prediction Consistency Criterion (LPCC) to find windows consistent with observed labels; Stage 2 applies confidence-based filtering with co-training (two networks exchange selected samples).
- Introduce MixAug, a time-series augmentation strategy for EEG, including Mixup-based augmentation with time-series transforms (e.g., Convolve/TimeWarp/Permutation).
Limitation
Our current evaluation is conducted on synthetically generated noisy datasets, which may not fully reflect real-world scenarios. We also restrict experiments to the subject-dependent setting, and future work aims to extend to subject-independent.
Abstract (from arXiv)
EEG-based emotion recognition has become an active research area in brain-computer interfaces. However, fine-grained affective annotation is costly, so most datasets provide only a global label for each long EEG trial. For practical emotion recognition, trials are commonly divided into short temporal windows, with the global label propagated to all windows for training. This practice treats trial-level emotion as temporally uniform, although affective responses may fluctuate within a trial. Consequently, windows that weakly reflect or deviate from the global emotional state may receive unreliable labels, posing a key challenge to robust EEG-based emotion recognition. To address this challenge, we propose VARY, a variability-aware robust learning framework for EEG-based emotion recognition from coarse trial-level labels. VARY incorporates a Two-stage Sample Selection (TSS) strategy to identify EEG windows with reliable emotional labels, and introduces MixAug, a data augmentation method tailored for EEG to improve model robustness under limited and unreliable supervision. We comprehensively evaluate VARY under the setting of label-noise learning on two public datasets, using three network architectures and multiple label-noise settings. Our method demonstrates superior performance, achieving a maximum accuracy gain of 11.54% on the DREAMER dataset and 5.39% on the SEED dataset compared with the best-performing baseline, highlighting the effectiveness of VARY in learning from unreliable trial-level labels under intra-trial affective variability.
Related papers
- Controlled Evaluation of EEG Foundation Models: Random Initialisation, Band-Power and Validation-Selected Probes
- InertiaMPDD: Multimodal Personalized Depression Detection via Emotional Inertia-Inspired Dynamics Modeling
- Similar Predictive Fit but Different Latent Dynamics: Characterizing Learned Dynamical Structure in Personalized Models of Brain Disorders
- TEX-MEX: Similarity-Guided Temporal Contrastive Learning for Time Series
- Your Vision Foundation Model Is Secretly an EEG Foundation Model
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First announced by arXiv: 2026-11-03 · Indexed by Pub-AI: 2026-11-03 UTC
Robust EEG-based Emotion Recognition from Coarse Trial-level Labels under Intra-trial Affective Variability
First announced by arXiv: 2026-11-03 · Original record ↗
- arxiv
- 2611.01731
Explicit source versions (1)
- v1 · Date not supplied
- Preprint available
Source date: 2026-11-03 · First detected: 2026-11-03 UTC · Historical baseline
Source evidence (1)
- Official evidence ↗ · RSS guid=oai:arXiv.org:2611.01731v1; arxiv:announce_type=new; item pubDate (official metadata)
Abstract · arxiv
EEG-based emotion recognition has become an active research area in brain-computer interfaces. However, fine-grained affective annotation is costly, so most datasets provide only a global label for each long EEG trial. For practical emotion recognition, trials are commonly divided into short temporal windows, with the global label propagated to all windows for training. This practice treats trial-level emotion as temporally uniform, although affective responses may fluctuate within a trial. Consequently, windows that weakly reflect or deviate from the global emotional state may receive unreliable labels, posing a key challenge to robust EEG-based emotion recognition. To address this challenge, we propose VARY, a variability-aware robust learning framework for EEG-based emotion recognition from coarse trial-level labels. VARY incorporates a Two-stage Sample Selection (TSS) strategy to identify EEG windows with reliable emotional labels, and introduces MixAug, a data augmentation method tailored for EEG to improve model robustness under limited and unreliable supervision. We comprehensively evaluate VARY under the setting of label-noise learning on two public datasets, using three network architectures and multiple label-noise settings. Our method demonstrates superior performance, achieving a maximum accuracy gain of 11.54% on the DREAMER dataset and 5.39% on the SEED dataset compared with the best-performing baseline, highlighting the effectiveness of VARY in learning from unreliable trial-level labels under intra-trial affective variability.
