Martin Uray

Zero-shot foundation models may excel at capturing temporal dynamics but fall short in reliably detecting multivariate anomalies, revealing a critical gap in their application.

Latent SDEs reveal a powerful new framework for detecting anomalies in multivariate time series, outperforming traditional methods in the face of data irregularities and sparsity.

Ditch the brittle Hough Transform: persistent homology offers a noise-resistant way to detect lines in point clouds.