CPCANet: Deep Unfolding Common Principal Component Analysis for Domain Generalization

Hugging Face Daily Papers Papers
domain-generalization deep-unfolding common-principal-component-analysis zero-shot-transfer out-of-distribution

Summary

CPCANet is a domain generalization framework that uses Common Principal Component Analysis to discover structured domain-invariant subspaces, achieving state-of-the-art performance in zero-shot transfer.

Domain Generalization (DG) aims to learn representations that remain robust under out-of-distribution (OOD) shifts and generalize effectively to unseen target domains. While recent invariant learning strategies and architectural advances have achieved strong performance, explicitly discovering a structured domain-invariant subspace through second-order statistics remains underexplored. In this work, we propose CPCANet, a novel framework grounded in Common Principal Component Analysis (CPCA), which unrolls the iterative Flury-Gautschi (FG) algorithm into fully differentiable neural layers. This approach integrates the statistical properties of CPCA into an end-to-end trainable framework, enforcing the discovery of a shared subspace across diverse domains while preserving interpretability. Experiments on four standard DG benchmarks demonstrate that CPCANet achieves state-of-the-art (SOTA) performance in zero-shot transfer. Moreover, CPCANet is architecture-agnostic and requires no dataset-specific tuning, providing a simple and efficient approach to learning robust representations under distribution shift. Code is available at https://github.com/wish44165/CPCANet.
Original Article
View Cached Full Text

Cached at: 05/11/26, 07:20 AM

Paper page - CPCANet: Deep Unfolding Common Principal Component Analysis for Domain Generalization

Source: https://huggingface.co/papers/2605.05136

Abstract

CPCANet is a domain generalization framework that uses Common Principal Component Analysis to discover structured domain-invariant subspaces through differentiable neural layers, achieving state-of-the-art performance in zero-shot transfer.

Domain Generalization(DG) aims to learn representations that remain robust under out-of-distribution (OOD) shifts and generalize effectively to unseen target domains. While recentinvariant learningstrategies and architectural advances have achieved strong performance, explicitly discovering a structureddomain-invariant subspacethrough second-order statistics remains underexplored. In this work, we propose CPCANet, a novel framework grounded inCommon Principal Component Analysis(CPCA), which unrolls the iterative Flury-Gautschi (FG) algorithm into fully differentiable neural layers. This approach integrates the statistical properties of CPCA into an end-to-end trainable framework, enforcing the discovery of a shared subspace across diverse domains while preserving interpretability. Experiments on four standard DG benchmarks demonstrate that CPCANet achieves state-of-the-art (SOTA) performance inzero-shot transfer. Moreover, CPCANet is architecture-agnostic and requires no dataset-specific tuning, providing a simple and efficient approach to learning robust representations under distribution shift. Code is available at https://github.com/wish44165/CPCANet.

View arXiv pageView PDFProject pageGitHub0Add to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.05136 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.05136 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.05136 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Projection Pursuit CPCANet for Domain Generalization

Hugging Face Daily Papers

Proposes PP-CPCANet, a covariance-free framework for domain generalization that learns a global orthogonal basis on the Stiefel manifold and achieves SOTA performance on four benchmarks.

Probe Generalization as Subspace Selection for OOD Deception Detection

arXiv cs.CL

This paper shows that projecting language model activations onto a small subset of principal components from the training distribution enables effective cross-domain transfer for deception detection probes, narrowing the performance gap between baseline and oracle methods.

Learning High Coverage Discriminative Parsimonious Rulesets

arXiv cs.LG

This paper introduces CDPR, a new approach for learning highly accurate and interpretable rule sets for classification using submodular maximization, achieving over 2.5x improvement in coverage compared to existing methods.

Hierarchical Domain Generalization

arXiv cs.LG

This paper introduces hierarchical domain generalization, formalizing extrapolation from finite observed regions to an entire instance space. It shows that no matter how simple the hypothesis class, certain domain partitions make generalization impossible, arguing that modern generalization theory must incorporate domain structure.