Learning Topology-Preserving Data Representations

01/31/2023
by   Ilya Trofimov, et al.
11

We propose a method for learning topology-preserving data representations (dimensionality reduction). The method aims to provide topological similarity between the data manifold and its latent representation via enforcing the similarity in topological features (clusters, loops, 2D voids, etc.) and their localization. The core of the method is the minimization of the Representation Topology Divergence (RTD) between original high-dimensional data and low-dimensional representation in latent space. RTD minimization provides closeness in topological features with strong theoretical guarantees. We develop a scheme for RTD differentiation and apply it as a loss term for the autoencoder. The proposed method “RTD-AE” better preserves the global structure and topology of the data manifold than state-of-the-art competitors as measured by linear correlation, triplet distance ranking accuracy, and Wasserstein distance between persistence barcodes.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/22/2018

Homology-Preserving Dimensionality Reduction via Manifold Landmarking and Tearing

Dimensionality reduction is an integral part of data visualization. It i...
research
01/31/2022

Topology-Preserving Dimensionality Reduction via Interleaving Optimization

Dimensionality reduction techniques are powerful tools for data preproce...
research
08/24/2023

Disentanglement Learning via Topology

We propose TopDis (Topological Disentanglement), a method for learning d...
research
04/23/2023

TopoSZ: Preserving Topology in Error-Bounded Lossy Compression

Existing error-bounded lossy compression techniques control the pointwis...
research
12/31/2021

Representation Topology Divergence: A Method for Comparing Neural Network Representations

Comparison of data representations is a complex multi-aspect problem tha...
research
03/05/2023

CAMEL: Curvature-Augmented Manifold Embedding and Learning

A novel method, named Curvature-Augmented Manifold Embedding and Learnin...
research
10/26/2021

Topologically penalized regression on manifolds

We study a regression problem on a compact manifold M. In order to take ...

Please sign up or login with your details

Forgot password? Click here to reset