Latent Manifold Reconstruction and Representation with Topological and Geometrical Regularization

TL;DR

This paper introduces an AutoEncoder-based approach that combines manifold reconstruction with topological and geometric regularizations to improve the discovery and preservation of latent manifold structures from noisy high-dimensional data.

Contribution

It proposes a novel method integrating manifold reconstruction and regularizations within an AutoEncoder to enhance manifold learning from noisy data.

Findings

Outperforms t-SNE, UMAP, and Topological AutoEncoders in manifold discovery.

Effectively preserves topological and geometric properties during dimensionality reduction.

Demonstrates robustness to noisy point cloud data.

Abstract

Manifold learning aims to discover and represent low-dimensional structures underlying high-dimensional data while preserving critical topological and geometric properties. Existing methods often fail to capture local details with global topological integrity from noisy data or construct a balanced dimensionality reduction, resulting in distorted or fractured embeddings. We present an AutoEncoder-based method that integrates a manifold reconstruction layer, which uncovers latent manifold structures from noisy point clouds, and further provides regularizations on topological and geometric properties during dimensionality reduction, whereas the two components promote each other during training. Experiments on point cloud datasets demonstrate that our method outperforms baselines like t-SNE, UMAP, and Topological AutoEncoders in discovering manifold structures from noisy data and preserving them through dimensionality reduction, as validated by visualization and quantitative metrics. This work demonstrates the significance of combining manifold reconstruction with manifold learning to achieve reliable representation of the latent manifold, particularly when dealing with noisy real-world data. Code repository: https://github.com/Thanatorika/mrtg.