Koopman Invariants as Drivers of Emergent Time-Series Clustering in Joint-Embedding Predictive Architectures

Pablo Ruiz-Morales, Dries Vanoost, Davy Pissoort +1

Joint-Embedding Predictive Architectures (JEPAs), a powerful class of self-supervised models, exhibit an unexplained ability to cluster time-series data by their underlying dynamic…