[Submitted on 6 May 2026 (v1), last revised 11 Nov 2026 (this version, v4)]
Abstract:Domain Generalization (DG) aims to learn representations that generalize to unseen target domains under out-of-distribution (OOD) shifts. Despite recent advances in invariant representation learning, explicitly capturing shared structures across source-domain distributions remains relatively underexplored. In this work, we integrate Common Principal Component Analysis (CPCA) into deep learning to identify a common orthogonal basis across multiple domains, proposing two complementary end-to-end differentiable frameworks. First, we introduce Common Principal Component Network (CPCNet), which unrolls the iterative Flury-Gautschi (FG) algorithm into fully differentiable neural layers, enabling covariance-based joint diagonalization within end-to-end representation learning. Second, to address rank deficiency arising from mini-batch covariance estimation in small-sample regimes, we develop Projection Pursuit-based Common Principal Component Network (PP-CPCNet), which directly optimizes a common orthogonal basis without explicit covariance estimation. Experiments validate the joint diagonalization capabilities of both methods, while evaluations on four standard DG benchmarks demonstrate competitive zero-shot generalization performance with favorable computational efficiency.
Submission history
From: Yu-Hsi Chen [view email]
[v1] Wed, 6 May 2026 17:09:34 UTC (49 KB)
[v2] Thu, 7 May 2026 06:48:04 UTC (51 KB)
[v3] Sat, 6 Jun 2026 14:23:47 UTC (51 KB)
[v4] Wed, 11 Nov 2026 04:03:54 UTC (11,348 KB)
