| title | Projection Pursuit CPCANet for Domain Generalization |
|---|---|
| authors | Yu-Hsi Chen, Abd-Krim Seghouane |
| date | 2026-07-24 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2607.22117 |
| affiliation | The University of Melbourne |
| links | abs pdf hf code |
| abstract | Domain Generalization (DG) aims to learn representations robust to distribution shifts. Recent geometric alignment methods, such as CPCANet, extract domain-invariant structures through batch-wise Common Principal Component Analysis (CPCA). However, CPCANet suffers from rank-deficient covariance estimation due to the small-sample-size issue in mini-batch training. To address this limitation, we propose Projection Pursuit CPCANet (PP-CPCANet), a covariance-free framework that learns a global orthogonal basis on the Stiefel manifold and jointly optimizes it with network parameters via the Cayley transform. We further introduce a symmetry-breaking detached-median PP dispersion objective to extract common principal components (CPCs) with dense and robust optimization signals. Experiments on four DG benchmarks show that PP-CPCANet achieves SOTA performance while maintaining stable training. |
cite
@misc{chen2026projection,
title={Projection Pursuit CPCANet for Domain Generalization},
author={Yu-Hsi Chen and Abd-Krim Seghouane},
year={2026},
eprint={2607.22117},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2607.22117}
}