Latest
xCEBRA: Time-series Attribution Maps with Regularized Contrastive Learning
xCEBRA generates attribution maps with identifiability guarantees for time-series data using regularized contrastive learning and a new …
Steffen Schneider, Rodrigo González Laiz, Anastasiia Filippova, Markus Frey, Mackenzie Weygandt Mathis
Self-supervised contrastive learning performs non-linear system identification
DCL is a framework for non-linear system identification using contrastive learning.