LGMLMar 1, 2019

Non-linear ICA based on Cramer-Wold metric

arXiv:1903.00201v16 citations
Originality Synthesis-oriented
AI Analysis

This work addresses non-linear source separation, an open problem with many applications, but it is incremental as it builds on a recently proposed method.

The paper tackled the challenging problem of non-linear source separation by introducing Cramer-Wold ICA (CW-ICA), which achieves comparable results to Adversarial Non-linear ICA (ANICA) while eliminating the need for adversarial training.

Non-linear source separation is a challenging open problem with many applications. We extend a recently proposed Adversarial Non-linear ICA (ANICA) model, and introduce Cramer-Wold ICA (CW-ICA). In contrast to ANICA we use a simple, closed--form optimization target instead of a discriminator--based independence measure. Our results show that CW-ICA achieves comparable results to ANICA, while foregoing the need for adversarial training.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes