Learning sleep stages from radio signals: A conditional adversarial architecture
Name
rfsleep-paper.pdf
Description
Accepted version
Size
590.2 KB
Format
Adobe PDF
Checksum (MD5)
bd817efbff16493abe358dff70cace04
Author(s)
Jaakkola, Tommi
Bianchi, Matt T.
Katabi, Dina
Yue, Shichao
Zhao, Mingmin
Date Issued
2017
Citation
Jaakkola, Tommi, Bianchi, Matt T., Katabi, Dina, Yue, Shichao and Zhao, Mingmin. 2017. "Learning sleep stages from radio signals: A conditional adversarial architecture."
Version
Author's final manuscript
Abstract
© Copyright 2017 by the authors(s). We focus on predicting sleep stages from radio measurements without any attached sensors on subjects. We introduce a new predictive model that combines convolutional and recurrent neural networks to extract sleep-specific subject-invariant features from RF signals and capture the temporal progression of sleep. A key innovation underlying our approach is a modified adversarial training regime that discards extraneous information specific to individuals or measurement conditions, while retaining all information relevant to the predictive task. We analyze our game theoretic setup and empirically demonstrate that our model achieves significant improvements over state-of-the-art solutions.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
http://proceedings.mlr.press/v70/zhao17d.html
