Latent Sensor Fusion: Multimedia Learning of Physiological Signals for Resource-Constrained Devices
Ahmed, Abdullah, Gummeson, Jeremy
–arXiv.org Artificial Intelligence
Latent spaces offer an efficient and effective means of summarizing data while implicitly preserving meta-information through relational encoding. We leverage these meta-embeddings to develop a modality-agnostic, unified encoder. Our method employs sensor-latent fusion to analyze and correlate multimodal physiological signals. Using a compressed sensing approach with autoencoder-based latent space fusion, we address the computational challenges of biosignal analysis on resource-constrained devices. Experimental results show that our unified encoder is significantly faster, lighter, and more scalable than modality-specific alternatives, without compromising representational accuracy.
arXiv.org Artificial Intelligence
Jul-22-2025
- Country:
- North America > United States > Massachusetts > Hampshire County > Amherst (0.15)
- Genre:
- Research Report (0.84)
- Industry:
- Health & Medicine > Diagnostic Medicine > Imaging (0.47)
- Technology:
- Information Technology
- Data Science (1.00)
- Artificial Intelligence
- Representation & Reasoning (1.00)
- Natural Language (1.00)
- Vision (0.94)
- Machine Learning > Neural Networks
- Deep Learning (0.95)
- Information Technology