Work

Time Series Source Separation with Slow Flows

Year: 2020

Type: preprint

Abstract: In this paper, we show that slow feature analysis (SFA), a common time series decomposition method, naturally fits into the flow-based models (FBM) framework, a type of invertible neural latent variable models. Building upon recent advances on blind source separation, we show that such a fit makes the time series decomposition identifiable.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2007.10182

Authors: Pineau, Edouard, Razakarivony, Sébastien, Bonald, Thomas

Institutions: Télécom Paris, Institut Polytechnique de Paris, Safran (France), Laboratoire Traitement et Communication de l’Information

Language: English

Topic: Blind Source Separation Techniques

Subfield: Signal Processing

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W3042874812

Cites: 24

Cited by:

Related to: 10