pith. sign in

arxiv: 2506.13714 · v1 · pith:XHAVVRZWnew · submitted 2025-06-16 · 📊 stat.ML · cs.LG· math.ST· stat.TH

Understanding Learning Invariance in Deep Linear Networks

classification 📊 stat.ML cs.LGmath.STstat.TH
keywords dataregularizationaugmentationinvariantlinearcriticaldeepglobal
0
0 comments X
read the original abstract

Equivariant and invariant machine learning models exploit symmetries and structural patterns in data to improve sample efficiency. While empirical studies suggest that data-driven methods such as regularization and data augmentation can perform comparably to explicitly invariant models, theoretical insights remain scarce. In this paper, we provide a theoretical comparison of three approaches for achieving invariance: data augmentation, regularization, and hard-wiring. We focus on mean squared error regression with deep linear networks, which parametrize rank-bounded linear maps and can be hard-wired to be invariant to specific group actions. We show that the critical points of the optimization problems for hard-wiring and data augmentation are identical, consisting solely of saddles and the global optimum. By contrast, regularization introduces additional critical points, though they remain saddles except for the global optimum. Moreover, we demonstrate that the regularization path is continuous and converges to the hard-wired solution.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Conservation Laws from Data Symmetry in Neural Networks

    cs.LG 2026-06 unverdicted novelty 7.0

    Data symmetries generically do not induce conserved quantities in NN training for analytic non-polynomial losses, but can for MSE with tensorizable networks.

  2. Equivariance and Augmentation for Bayesian Neural Networks

    cs.LG 2026-06 unverdicted novelty 5.0

    Derives exact equivariance conditions for augmented BNNs under variational inference and proposes orbit expansion symmetrization that outperforms baselines on equivariance and accuracy.