analiza aiqbench
This research introduces Phenomenon-Graph JEPA, a joint-embedding predictive architecture designed to improve label efficiency in contactless cardiorespiratory sensing using mmWave radar and RGB-D cameras. By leveraging self-supervised pretraining on unlabeled signals without relying on negative pairs or synthetic augmentation, the model demonstrates a measurable gain in label-efficiency area compared to matched supervised training on the OMuSense-23 dataset. The study isolates the specific contribution of predictive representations from physiological priors, offering a more robust approach for handling scarce labeled data in continuous health monitoring scenarios.
Phenomenon-Graph JEPA: Label-Efficient Representation Learning for Contactless Cardiorespiratory Sensing
arXiv:2609.32928v1 Announce Type: new Abstract: Millimeter-wave (mmWave) radar and RGB-D cameras can record cardiac and respiratory waveforms continuously and without contact, but labeled recordings remain scarce because every label requires a supervised acquisition session. Self-supervised pretraining can exploit the unlabeled signals, yet contrastive methods depend on signal transformations and negative pairs whose validity is uncertain for cardiorespiratory data, where time warping changes breathing rate and distant windows can share the same physiological state. We present Phenomenon-Graph JEPA, a joint-embedding predictive architecture that learns from four processed one-dimensional streams without negative pairs or synthetic augmentation in its base configuration. Each stream is encoded by a temporal convolutional branch and a band-limited spectral branch. During pretraining, the model predicts stopped target embeddings along typed edges, which connect streams assigned to the same physiological phenomenon, and forward in time within a state episode. We treat this physiological typing as a testable hypothesis and compare it with wrong-edge and all-pairs prediction graphs. In the OMuSense-23 dataset, pretraining improves label-efficiency area over matched supervised training by 3.91 percentage points (95% interval 2.08 to 5.80, Holm-adjusted p = 0.006), and by 3.74 points under a second configuration evaluated on the same test participants. However, the wrong-edge and all-pairs controls do not establish a benefit from physiological typing. Optional Takens-inspired delay coordinates improve a validation comparison with learned history, whereas two wrist-only WESAD protocols do not establish a pretraining advantage. The study therefore separates the measured benefit of predictive representations from the physiological prior used to organize their training.
Abstract:Millimeter-wave (mmWave) radar and RGB-D cameras can record cardiac and respiratory waveforms continuously and without contact, but labeled recordings remain scarce because every label requires a supervised acquisition session. Self-supervised pretraining can exploit the unlabeled signals, yet contrastive methods depend on signal transformations and negative pairs whose validity is uncertain for cardiorespiratory data, where time warping changes breathing rate and distant windows can share the same physiological state. We present Phenomenon-Graph JEPA, a joint-embedding predictive architecture that learns from four processed one-dimensional streams without negative pairs or synthetic augmentation in its base configuration. Each stream is encoded by a temporal convolutional branch and a band-limited spectral branch. During pretraining, the model predicts stopped target embeddings along typed edges, which connect streams assigned to the same physiological phenomenon, and forward in time within a state episode. We treat this physiological typing as a testable hypothesis and compare it with wrong-edge and all-pairs prediction graphs. In the OMuSense-23 dataset, pretraining improves label-efficiency area over matched supervised training by 3.91 percentage points (95% interval 2.08 to 5.80, Holm-adjusted p = 0.006), and by 3.74 points under a second configuration evaluated on the same test participants. However, the wrong-edge and all-pairs controls do not establish a benefit from physiological typing. Optional Takens-inspired delay coordinates improve a validation comparison with learned history, whereas two wrist-only WESAD protocols do not establish a pretraining advantage. The study therefore separates the measured benefit of predictive representations from the physiological prior used to organize their training.
9 pages, 4 figures, 7 tables, 24 numbered equations. Extended version of a conference submission, with additional methodological details, implementation settings, and analyzes
Machine Learning (cs.LG); Emerging Technologies (cs.ET)
arXiv:2609.32928 [cs.LG]
(or arXiv:2609.32928v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.32928 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 291 cuvinte
Subiecte conexe
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · 8 oct. 2026
The J-Space Debate, Agent Swarms, and Pacing Frontier AI - Digital Minds Newsletter #4
LessWrong · 18 sept. 2026
Robust and Efficient AI Frameworks for Scalable Material Design and Property Prediction
arXiv · 17 sept. 2026
World Models and Yann LeCun’s Bet Against Large Language Models
Quantum Zeitgeist · 17 iul. 2026