Reproduce “Exploring Generalist Foundation Models for Time Series of Electrodermal Activity Data” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Leonardo Alchieri, Lino Candian, Nouran Abdalazim, Lidia Alecci, Giovanni De Felice, Silvia Santini
The authors test whether pretrained time-series models provide useful features for wearable electrodermal activity (EDA) classification. They compare four models’ frozen representations with handcrafted features using logistic-regression probes and leave-one-participant-out evaluation across four tasks. On USILaughs, Chronos Large reaches 0.72 ± 0.07 balanced accuracy versus 0.73 ± 0.09 for handcrafted features; no foundation model significantly outperforms that baseline across the tasks.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.