Unit-Consistent (UC) Adjoint for GSD and Backprop in Deep Learning Applications
Abstract
Deep neural networks constructed from linear maps and positively homogeneous nonlinearities (e.g., ReLU) possess a fundamental gauge symmetry: the network function is invariant to node-wise diagonal rescalings. However, standard gradient descent is not equivariant to this symmetry, causing optimization trajectories to depend heavily on arbitrary parameterizations. Prior work has proposed rescaling-invariant optimization schemes for positively homogeneous networks (e.g., path-based or path-space updates). Our contribution is complementary: we formulate the invariance requirement at the level of the backward adjoint/optimization geometry, which provides a simple, operator-level recipe that can be applied uniformly across network components and optimizer state. By replacing the Euclidean transpose with a Unit-Consistent (UC) adjoint, we derive UC gauge-consistent steepest descent and backprogation.
Cite
@article{arxiv.2601.10873,
title = {Unit-Consistent (UC) Adjoint for GSD and Backprop in Deep Learning Applications},
author = {Jeffrey Uhlmann},
journal= {arXiv preprint arXiv:2601.10873},
year = {2026}
}