aiqbench analysis
This research advances self-supervised learning for human activity recognition by adapting the Joint Embedding Predictive Architecture to sensor data, effectively reducing reliance on costly labeled datasets. By integrating a modified VICReg regularization objective, the framework stabilizes pre-training and achieves superior generalization in complex transitional movements like sit-to-stand transitions where supervised models typically overfit. The work demonstrates that robust temporal representations can be learned from unlabeled streams of sensor inputs without sacrificing performance on minority activity classes.
Joint-Embedding Predictive Architecture for Sensor-based Activity Recognition
arXiv:2607.16350v1 Announce Type: cross Abstract: Sensor-based human activity recognition (HAR) has achieved significant progressed in fully supervised learning settings. However, these supervised learning models rely on large amount of labeled data, which require labor-intensive collection and meticulous annotation. To address these challenges, this paper proposes a Joint Embedding Predictive Architecture framework tailored for sensor-based HAR, designed to learn robust and generalizable representations from unlabeled datasets. The proposed framework features an encoder designed to explicitly model both the fine-grained local temporal representations within individual window and the long-term temporal sequence of adjacent windows. Furthermore, we introduce an improved Variance-Invariance-Covariance Regularization (VICReg) objective function that incorporates computationally lightweight norm term to stabilize the JEPA pre-training phase. This term balances variance, invariance and covariance constraints to prevent representation collapse. The proposed HAR-JEPA framework is evaluated using two benchmark continuously performed activity datasets. The results show that high-quality representations are successfully learned by the proposed framework. Furthermore, the representations learned by HAR-JEPA demonstrates superior generalization on minority, high variance transitional activities such as sit-to-stand and sit-to-lie where supervised learning tend to overfit due to limited support.
Abstract:Sensor-based human activity recognition (HAR) has achieved significant progressed in fully supervised learning settings. However, these supervised learning models rely on large amount of labeled data, which require labor-intensive collection and meticulous annotation. To address these challenges, this paper proposes a Joint Embedding Predictive Architecture framework tailored for sensor-based HAR, designed to learn robust and generalizable representations from unlabeled datasets. The proposed framework features an encoder designed to explicitly model both the fine-grained local temporal representations within individual window and the long-term temporal sequence of adjacent windows. Furthermore, we introduce an improved Variance-Invariance-Covariance Regularization (VICReg) objective function that incorporates computationally lightweight norm term to stabilize the JEPA pre-training phase. This term balances variance, invariance and covariance constraints to prevent representation collapse. The proposed HAR-JEPA framework is evaluated using two benchmark continuously performed activity datasets. The results show that high-quality representations are successfully learned by the proposed framework. Furthermore, the representations learned by HAR-JEPA demonstrates superior generalization on minority, high variance transitional activities such as sit-to-stand and sit-to-lie where supervised learning tend to overfit due to limited support.
Signal Processing (eess.SP); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2607.16350 [eess.SP]
(or arXiv:2607.16350v1 [eess.SP] for this version)
https://doi.org/10.48550/arXiv.2607.16350 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 277 words
Related coverage
News
Quantum Computing Report · Jul 25, 2026
W-JEPA: Wavelet Joint-Embedding Predictive Architecture for Wearable Activity Recognition
arXiv · Oct 15, 2026
Databricks announces 2026 global partner awards
Databricks · Jun 15, 2026
News Archive 2026
Quantum Computing Report · Jul 25, 2026