{"slug": "explainable-graph-attention-network-for-stress-recognition-stressgat-via-action", "title": "Explainable graph attention network for stress recognition (StressGAT) via differential action units", "summary": "Researchers introduce StressGAT, a Graph Attention Network that achieves 88.62% accuracy in stress recognition by using Differential Action Units to normalize individual facial expressions relative to neutral baselines. The model, tested on 58 participants via Leave-One-Subject-Out cross-validation, integrates a Multiple Instance Learning attention mechanism to identify peak stress intervals and reveal distinct expressivity phenotypes, offering an explainable solution for personalized affective monitoring.", "body_md": "arXiv:2607.20819v1 Announce Type: new\nAbstract: Stress is a dynamic process characterized by significant individual variability in facial expression. Traditional architectures, such as Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), often overlook person-specific baselines or lack the representational capacity to model the non-linear temporal progression of distress due to sequential bottlenecks and rigid grid-based constraints. Furthermore, many deep learning models lack the interpretability required for clinical deployment. This study introduces StressGAT, a Graph Attention Network that leverages the relational inductive bias of graph modeling to capture complex facial dynamics that indicate acute stress. By using Differential Action Units, the framework normalizes individual responses relative to neutral baselines to achieve personalized recognition. The proposed model achieves 88.62\\% accuracy on a diverse stress-induction cohort (58 participants) using a subject-independent, Leave-One-Subject-Out (LOSO) cross-validation protocol. Beyond predictive accuracy, the architecture integrates a Multiple Instance Learning (MIL) attention mechanism to identify peak stress intervals and reveal distinct expressivity phenotypes. By simultaneously optimizing for accuracy and interpretability, this framework provides a robust, explainable solution for personalized affective monitoring.", "url": "https://wpnews.pro/news/explainable-graph-attention-network-for-stress-recognition-stressgat-via-action", "canonical_source": "https://arxiv.org/abs/2607.20819", "published_at": "2026-07-24 04:00:00+00:00", "updated_at": "2026-07-24 04:28:32.288564+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "neural-networks", "ai-research"], "entities": ["StressGAT", "Graph Attention Network", "Multiple Instance Learning", "Differential Action Units", "Leave-One-Subject-Out"], "alternates": {"html": "https://wpnews.pro/news/explainable-graph-attention-network-for-stress-recognition-stressgat-via-action", "markdown": "https://wpnews.pro/news/explainable-graph-attention-network-for-stress-recognition-stressgat-via-action.md", "text": "https://wpnews.pro/news/explainable-graph-attention-network-for-stress-recognition-stressgat-via-action.txt", "jsonld": "https://wpnews.pro/news/explainable-graph-attention-network-for-stress-recognition-stressgat-via-action.jsonld"}}