Towards The Implicit Bias on Multiclass Separable Data Under Norm Constraints

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xie, Shengping, Wu, Zekun, Chen, Quan, Tang, Kaixu
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914417129553920
author Xie, Shengping
Wu, Zekun
Chen, Quan
Tang, Kaixu
author_facet Xie, Shengping
Wu, Zekun
Chen, Quan
Tang, Kaixu
contents Implicit bias induced by gradient-based algorithms is essential to the generalization of overparameterized models, yet its mechanisms can be subtle. This work leverages the Normalized Steepest Descent} (NSD) framework to investigate how optimization geometry shapes solutions on multiclass separable data. We introduce NucGD, a geometry-aware optimizer designed to enforce low rank structures through nuclear norm constraints. Beyond the algorithm itself, we connect NucGD with emerging low-rank projection methods, providing a unified perspective. To enable scalable training, we derive an efficient SVD-free update rule via asynchronous power iteration. Furthermore, we empirically dissect the impact of stochastic optimization dynamics, characterizing how varying levels of gradient noise induced by mini-batch sampling and momentum modulate the convergence toward the expected maximum margin solutions.Our code is accessible at: https://github.com/Tsokarsic/observing-the-implicit-bias-on-multiclass-seperable-data.
format Preprint
id arxiv_https___arxiv_org_abs_2603_22824
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Towards The Implicit Bias on Multiclass Separable Data Under Norm Constraints
Xie, Shengping
Wu, Zekun
Chen, Quan
Tang, Kaixu
Machine Learning
Optimization and Control
Implicit bias induced by gradient-based algorithms is essential to the generalization of overparameterized models, yet its mechanisms can be subtle. This work leverages the Normalized Steepest Descent} (NSD) framework to investigate how optimization geometry shapes solutions on multiclass separable data. We introduce NucGD, a geometry-aware optimizer designed to enforce low rank structures through nuclear norm constraints. Beyond the algorithm itself, we connect NucGD with emerging low-rank projection methods, providing a unified perspective. To enable scalable training, we derive an efficient SVD-free update rule via asynchronous power iteration. Furthermore, we empirically dissect the impact of stochastic optimization dynamics, characterizing how varying levels of gradient noise induced by mini-batch sampling and momentum modulate the convergence toward the expected maximum margin solutions.Our code is accessible at: https://github.com/Tsokarsic/observing-the-implicit-bias-on-multiclass-seperable-data.
title Towards The Implicit Bias on Multiclass Separable Data Under Norm Constraints
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2603.22824