Steven M. Peterson

This paper develops cross-modal, self-supervised neural decoders using multiple data streams to achieve high decoding accuracy without labeled data.

AJILE12 provides a large dataset of intracranial neural recordings and upper body pose data from 12 participants over 55 days, with metadata on movement and behavior.

The paper analyzes neural variability during naturalistic arm movements using intracranial recordings and machine learning, finding significant variability across subjects and events.

This paper introduces HTNet, a neural decoder that generalizes across participants and recording modalities, achieving performance comparable to tailored decoders with minimal data.