pith. sign in

arxiv: 2104.03821 · v1 · pith:AGSTTL7Xnew · submitted 2021-04-08 · 💻 cs.CV

Robust Differentiable SVD

classification 💻 cs.CV
keywords computeeigendecompositiongradientsduringeigenvectorsgradienthoweveriterative
0
0 comments X
read the original abstract

Eigendecomposition of symmetric matrices is at the heart of many computer vision algorithms. However, the derivatives of the eigenvectors tend to be numerically unstable, whether using the SVD to compute them analytically or using the Power Iteration (PI) method to approximate them. This instability arises in the presence of eigenvalues that are close to each other. This makes integrating eigendecomposition into deep networks difficult and often results in poor convergence, particularly when dealing with large matrices. While this can be mitigated by partitioning the data into small arbitrary groups, doing so has no theoretical basis and makes it impossible to exploit the full power of eigendecomposition. In previous work, we mitigated this using SVD during the forward pass and PI to compute the gradients during the backward pass. However, the iterative deflation procedure required to compute multiple eigenvectors using PI tends to accumulate errors and yield inaccurate gradients. Here, we show that the Taylor expansion of the SVD gradient is theoretically equivalent to the gradient obtained using PI without relying in practice on an iterative process and thus yields more accurate gradients. We demonstrate the benefits of this increased accuracy for image classification and style transfer.

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 1 Pith paper

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

  1. Computation-Aware Kalman Filtering with Model Selection for Neural Dynamics

    stat.ML 2026-05 unverdicted novelty 5.0

    Introduces CASSM, a computation-aware state-space model extending Kalman filtering with model selection for scale-imbalanced neural recordings, claiming competitive performance with deep networks and improved uncertai...