Comparative Analysis in Pre-image Algorithms of Kernel PCA
Wojciech Czaja, Canran Ji
Proceedings of the 1st Conference on Topology, Algebra, and Geometry in Data Science(TAG-DS 2025), PMLR 321:403-413, 2026.
Abstract
We study the kernel PCA (kPCA) pre-image problem in image denoising by benchmarking classical algorithms and introducing two neural network adversarial pre-imaging models, DCGAN-KPCAnet and WGAN-KPCAnet. Our results show that WGAN-KPCAnet delivers superior reconstruction results and is robust to noise compared to baselines.
Cite this Paper
BibTeX
@InProceedings{pmlr-v321-czaja26a,
title = {Comparative Analysis in Pre-image Algorithms of Kernel PCA},
author = {Czaja, Wojciech and Ji, Canran},
booktitle = {Proceedings of the 1st Conference on Topology, Algebra, and Geometry in Data Science(TAG-DS 2025)},
pages = {403--413},
year = {2026},
editor = {Bernardez Gil, Guillermo and Black, Mitchell and Cloninger, Alexander and Doster, Timothy and Emerson, Tegan and Garcı́a-Rodondo, Ińes and Holtz, Chester and Kotak, Mit and Kvinge, Henry and Mishne, Gal and Papillon, Mathilde and Pouplin, Alison and Rainey, Katie and Rieck, Bastian and Telyatnikov, Lev and Yeats, Eric and Wang, Qingsong and Wang, Yusu and Wayland, Jeremy},
volume = {321},
series = {Proceedings of Machine Learning Research},
month = {01--02 Dec},
publisher = {PMLR},
pdf = {https://raw.githubusercontent.com/mlresearch/v321/main/assets/czaja26a/czaja26a.pdf},
url = {https://proceedings.mlr.press/v321/czaja26a.html},
abstract = {We study the kernel PCA (kPCA) pre-image problem in image denoising by benchmarking classical algorithms and introducing two neural network adversarial pre-imaging models, DCGAN-KPCAnet and WGAN-KPCAnet. Our results show that WGAN-KPCAnet delivers superior reconstruction results and is robust to noise compared to baselines.}
}
Endnote
%0 Conference Paper
%T Comparative Analysis in Pre-image Algorithms of Kernel PCA
%A Wojciech Czaja
%A Canran Ji
%B Proceedings of the 1st Conference on Topology, Algebra, and Geometry in Data Science(TAG-DS 2025)
%C Proceedings of Machine Learning Research
%D 2026
%E Guillermo Bernardez Gil
%E Mitchell Black
%E Alexander Cloninger
%E Timothy Doster
%E Tegan Emerson
%E Ińes Garcı́a-Rodondo
%E Chester Holtz
%E Mit Kotak
%E Henry Kvinge
%E Gal Mishne
%E Mathilde Papillon
%E Alison Pouplin
%E Katie Rainey
%E Bastian Rieck
%E Lev Telyatnikov
%E Eric Yeats
%E Qingsong Wang
%E Yusu Wang
%E Jeremy Wayland
%F pmlr-v321-czaja26a
%I PMLR
%P 403--413
%U https://proceedings.mlr.press/v321/czaja26a.html
%V 321
%X We study the kernel PCA (kPCA) pre-image problem in image denoising by benchmarking classical algorithms and introducing two neural network adversarial pre-imaging models, DCGAN-KPCAnet and WGAN-KPCAnet. Our results show that WGAN-KPCAnet delivers superior reconstruction results and is robust to noise compared to baselines.
APA
Czaja, W. & Ji, C.. (2026). Comparative Analysis in Pre-image Algorithms of Kernel PCA. Proceedings of the 1st Conference on Topology, Algebra, and Geometry in Data Science(TAG-DS 2025), in Proceedings of Machine Learning Research 321:403-413 Available from https://proceedings.mlr.press/v321/czaja26a.html.
![[Topology, Algebra, and Geometry in Data Science(TAG-DS 2025) Logo]](/v321/assets/images/logo-pmlr.svg){kind=logo}