Skip to main content
. Author manuscript; available in PMC: 2023 Sep 12.
Published in final edited form as: IEEE J Biomed Health Inform. 2023 Feb 3;27(2):1106–1117. doi: 10.1109/JBHI.2022.3224727

Fig. 2. Illustration of BYOL and BYOL for hierarchical BEHRT.

Fig. 2

BYOL (top) includes four components, data augmentation, encoder, projector, and predictor. Both online and target network have a similar model architecture. BYOL for hierarchical BEHRT (bottom) has similar framework, the segment augmentation corresponds to the augmentation in BYOL (top), and the segment augmentation is conducted on temporal features extracted by local feature extractor.

HHS Vulnerability Disclosure