Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Vigna, Hugo, Bontemps, Samuel
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.07790
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916019868532736
author Vigna, Hugo
Bontemps, Samuel
author_facet Vigna, Hugo
Bontemps, Samuel
contents The Hessian spectrum of trained deep networks exhibits a characteristic structure: a continuous bulk of near-zero eigenvalues and a small number of large outlier eigenvalues (spikes), confirming the relevance of Random Matrix Theory in deep learning. The spike count matches the number of classes minus one. While prior work has described this structure, no method has exploited it operationally to improve classification performance. We propose Hessian Surgery, a post-hoc optimization method that directly perturbs model weights along spike eigenvectors to rebalance per-class accuracy without retraining. We introduce (i) a spike-class sensitivity matrix that quantifies the directional derivative of each class's accuracy along each spike eigenvector, (ii) a constrained optimization of perturbation coefficients that targets weak classes while preserving strong ones, and (iii) an adaptive amplitude control that raises or lowers the perturbation budget based on iteration-level improvement signals. We obtain encouraging results on CIFAR-10 and ISIC-2019 on both balanced accuracy and standard deviation.
format Preprint
id arxiv_https___arxiv_org_abs_2605_07790
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Hessian Surgery: Class-Targeted Post-Hoc Rebalancing via Hessian Spike Perturbation
Vigna, Hugo
Bontemps, Samuel
Machine Learning
Computer Vision and Pattern Recognition
The Hessian spectrum of trained deep networks exhibits a characteristic structure: a continuous bulk of near-zero eigenvalues and a small number of large outlier eigenvalues (spikes), confirming the relevance of Random Matrix Theory in deep learning. The spike count matches the number of classes minus one. While prior work has described this structure, no method has exploited it operationally to improve classification performance. We propose Hessian Surgery, a post-hoc optimization method that directly perturbs model weights along spike eigenvectors to rebalance per-class accuracy without retraining. We introduce (i) a spike-class sensitivity matrix that quantifies the directional derivative of each class's accuracy along each spike eigenvector, (ii) a constrained optimization of perturbation coefficients that targets weak classes while preserving strong ones, and (iii) an adaptive amplitude control that raises or lowers the perturbation budget based on iteration-level improvement signals. We obtain encouraging results on CIFAR-10 and ISIC-2019 on both balanced accuracy and standard deviation.
title Hessian Surgery: Class-Targeted Post-Hoc Rebalancing via Hessian Spike Perturbation
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.07790