Saved in:
Bibliographic Details
Main Authors: Li, Bingcong, Zhang, Yilang, Giannakis, Georgios B.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.02433
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909766736936960
author Li, Bingcong
Zhang, Yilang
Giannakis, Georgios B.
author_facet Li, Bingcong
Zhang, Yilang
Giannakis, Georgios B.
contents Sharpness-aware minimization (SAM) has well-documented merits in enhancing generalization of deep neural network models. Accounting for sharpness in the loss function geometry, where neighborhoods of `flat minima' heighten generalization ability, SAM seeks `flat valleys' by minimizing the maximum loss provoked by an adversarial perturbation within the neighborhood. Although critical to account for sharpness of the loss function, in practice SAM suffers from `over-friendly adversaries,' which can curtail the outmost level of generalization. To avoid such `friendliness,' the present contribution fosters stabilization of adversaries through variance suppression (VASSO). VASSO offers a general approach to provably stabilize adversaries. In particular, when integrating VASSO with SAM, improved generalizability is numerically validated on extensive vision and language tasks. Once applied on top of a computationally efficient SAM variant, VASSO offers a desirable generalization-computation tradeoff.
format Preprint
id arxiv_https___arxiv_org_abs_2509_02433
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle VASSO: Variance Suppression for Sharpness-Aware Minimization
Li, Bingcong
Zhang, Yilang
Giannakis, Georgios B.
Machine Learning
Sharpness-aware minimization (SAM) has well-documented merits in enhancing generalization of deep neural network models. Accounting for sharpness in the loss function geometry, where neighborhoods of `flat minima' heighten generalization ability, SAM seeks `flat valleys' by minimizing the maximum loss provoked by an adversarial perturbation within the neighborhood. Although critical to account for sharpness of the loss function, in practice SAM suffers from `over-friendly adversaries,' which can curtail the outmost level of generalization. To avoid such `friendliness,' the present contribution fosters stabilization of adversaries through variance suppression (VASSO). VASSO offers a general approach to provably stabilize adversaries. In particular, when integrating VASSO with SAM, improved generalizability is numerically validated on extensive vision and language tasks. Once applied on top of a computationally efficient SAM variant, VASSO offers a desirable generalization-computation tradeoff.
title VASSO: Variance Suppression for Sharpness-Aware Minimization
topic Machine Learning
url https://arxiv.org/abs/2509.02433