Trace your sources in large-scale data: one ring to find them all

TL;DR

This paper introduces DECOMPOSE, a scalable probabilistic framework for blind source separation that improves accuracy and robustness over traditional methods, effectively handling large-scale data in data analysis pipelines.

Contribution

The paper presents DECOMPOSE, a novel probabilistic BSS framework that generalizes traditional algorithms and offers improved accuracy, robustness, and scalability for large-scale data analysis.

Findings

Substantial improvements in accuracy over traditional BSS methods.

Enhanced robustness and reliability in source extraction.

Effective handling of large-scale data with algorithmic efficiency.

Abstract

An important preprocessing step in most data analysis pipelines aims to extract a small set of sources that explain most of the data. Currently used algorithms for blind source separation (BSS), however, often fail to extract the desired sources and need extensive cross-validation. In contrast, their rarely used probabilistic counterparts can get away with little cross-validation and are more accurate and reliable but no simple and scalable implementations are available. Here we present a novel probabilistic BSS framework (DECOMPOSE) that can be flexibly adjusted to the data, is extensible and easy to use, adapts to individual sources and handles large-scale data through algorithmic efficiency. DECOMPOSE encompasses and generalises many traditional BSS algorithms such as PCA, ICA and NMF and we demonstrate substantial improvements in accuracy and robustness on artificial and real data.