Saved in:
Bibliographic Details
Main Authors: Galperin, Daniel, Köthe, Ullrich
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.06940
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910014381228032
author Galperin, Daniel
Köthe, Ullrich
author_facet Galperin, Daniel
Köthe, Ullrich
contents Learning unsupervised representations that are both semantically meaningful and stable across runs remains a central challenge in modern representation learning. We introduce entropy-ordered flows (EOFlows), a normalizing-flow framework that orders latent dimensions by their explained entropy, analogously to PCA's explained variance. This ordering enables adaptive injective flows: after training, one may retain only the top C latent variables to form a compact core representation while the remaining variables capture fine-grained detail and noise, with C chosen flexibly at inference time rather than fixed during training. EOFlows build on insights from Independent Mechanism Analysis, Principal Component Flows and Manifold Entropic Metrics. We combine likelihood-based training with local Jacobian regularization and noise augmentation into a method that scales well to high-dimensional data such as images. Experiments on the CelebA dataset show that our method uncovers a rich set of semantically interpretable features, allowing for high compression and strong denoising.
format Preprint
id arxiv_https___arxiv_org_abs_2602_06940
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Core to Detail: Unsupervised Disentanglement with Entropy-Ordered Flows
Galperin, Daniel
Köthe, Ullrich
Machine Learning
Learning unsupervised representations that are both semantically meaningful and stable across runs remains a central challenge in modern representation learning. We introduce entropy-ordered flows (EOFlows), a normalizing-flow framework that orders latent dimensions by their explained entropy, analogously to PCA's explained variance. This ordering enables adaptive injective flows: after training, one may retain only the top C latent variables to form a compact core representation while the remaining variables capture fine-grained detail and noise, with C chosen flexibly at inference time rather than fixed during training. EOFlows build on insights from Independent Mechanism Analysis, Principal Component Flows and Manifold Entropic Metrics. We combine likelihood-based training with local Jacobian regularization and noise augmentation into a method that scales well to high-dimensional data such as images. Experiments on the CelebA dataset show that our method uncovers a rich set of semantically interpretable features, allowing for high compression and strong denoising.
title From Core to Detail: Unsupervised Disentanglement with Entropy-Ordered Flows
topic Machine Learning
url https://arxiv.org/abs/2602.06940