pith. sign in

arxiv: 2507.18807 · v1 · pith:CYKS52THnew · submitted 2025-07-24 · 💻 cs.LG

Fishers for Free? Approximating the Fisher Information Matrix by Recycling the Squared Gradient Accumulator

classification 💻 cs.LG
keywords fisherdiagonalgradientsquaredaccumulatorapproximationbeencourse
0
0 comments X
read the original abstract

The diagonal of a model's Fisher Information Matrix (the "Fisher diagonal") has frequently been used as a way to measure parameter sensitivity. Typically, the Fisher diagonal is estimated via squared sampled gradients of the model's likelihood with respect to its parameters, averaged over a few hundred or thousand examples -- a process which incurs nontrivial computational costs. At the same time, adaptive gradient methods like the ubiquitous Adam optimizer compute a moving average of the squared gradient over the course of training. This paper therefore explores whether an approximation of the Fisher diagonal can be obtained "for free" by recycling the squared gradient accumulator that has already been computed over the course of training. Through a comprehensive set of experiments covering five applications of the Fisher diagonal, we demonstrate that the "Squisher" (SQUared gradient accumulator as an approximation of the FISHER) consistently performs similarly to the Fisher diagonal while outperforming baseline methods. Additionally, we clarify the exact differences between the Squisher and the Fisher diagonal and provide empirical quantification of their respective impact.

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. MAdam: Metric-Aware Multi-Objective Adam

    cs.LG 2026-06 unverdicted novelty 7.0

    MAdam preconditions MOO solver directions with preference-conditioned curvature so that Adam's adaptive steps respect the intended metric instead of entangling it with gradient history.