GT-PCA: Effective and Interpretable Dimensionality Reduction with General Transform-Invariant Principal Component Analysis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Heinrichs, Florian
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929226250190848
author Heinrichs, Florian
author_facet Heinrichs, Florian
contents Data analysis often requires methods that are invariant with respect to specific transformations, such as rotations in case of images or shifts in case of images and time series. While principal component analysis (PCA) is a widely-used dimension reduction technique, it lacks robustness with respect to these transformations. Modern alternatives, such as autoencoders, can be invariant with respect to specific transformations but are generally not interpretable. We introduce General Transform-Invariant Principal Component Analysis (GT-PCA) as an effective and interpretable alternative to PCA and autoencoders. We propose a neural network that efficiently estimates the components and show that GT-PCA significantly outperforms alternative methods in experiments based on synthetic and real data.
format Preprint
id arxiv_https___arxiv_org_abs_2401_15623
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle GT-PCA: Effective and Interpretable Dimensionality Reduction with General Transform-Invariant Principal Component Analysis
Heinrichs, Florian
Machine Learning
Methodology
62H25 (Primary) 62M10, 62R10, 68T07, 68T10, 62M45 (Secondary)
G.3; I.2.6; I.5.1
Data analysis often requires methods that are invariant with respect to specific transformations, such as rotations in case of images or shifts in case of images and time series. While principal component analysis (PCA) is a widely-used dimension reduction technique, it lacks robustness with respect to these transformations. Modern alternatives, such as autoencoders, can be invariant with respect to specific transformations but are generally not interpretable. We introduce General Transform-Invariant Principal Component Analysis (GT-PCA) as an effective and interpretable alternative to PCA and autoencoders. We propose a neural network that efficiently estimates the components and show that GT-PCA significantly outperforms alternative methods in experiments based on synthetic and real data.
title GT-PCA: Effective and Interpretable Dimensionality Reduction with General Transform-Invariant Principal Component Analysis
topic Machine Learning
Methodology
62H25 (Primary) 62M10, 62R10, 68T07, 68T10, 62M45 (Secondary)
G.3; I.2.6; I.5.1
url https://arxiv.org/abs/2401.15623