paper-with-me

홈 › Papers

Rethinking Gauss-Newton for learning over-parameterized models

2023-02-06 · NeurIPS 2023 11 · Michael Arbel, Romain Menegaux, Pierre Wolinski

This work studies the global convergence and implicit bias of Gauss Newton's (GN) when optimizing over-parameterized one-hidden layer networks in the mean-field regime. We first establish a global convergence result for GN in the continuous-time limit exhibiting a faster convergence rate compared to GD due to improved conditioning. We then perform an empirical study on a synthetic regression task to investigate the implicit bias of GN's method. While GN is consistently faster than GD in finding a global optimum, the learned model generalizes well on test data when starting from random initial weights with a small variance and using a small step size to slow down convergence. Specifically, our study shows that such a setting results in a hidden learning phenomenon, where the dynamics are able to recover features with good generalization properties despite the model having sub-optimal training and test performances due to an under-optimized linear layer. This study exhibits a trade-off between the convergence speed of GN and the generalization ability of the learned solution.

📄 PDF Abstract BibTeX arXiv:2302.02904

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…

Similar Papers 제목 키워드 기반

Non-Asymptotic Optimization and Generalization Bounds for Stochastic Gauss-Newton in Overparameterized Models

2025-11-06 · Semih Cayci arxiv

An important question in deep learning is how higher-order optimization methods affect generalization. In this work, we analyze a stochastic Gauss-Newton (SGN) method with Levenberg-Marquardt damping and mini-batch sampl…

A Riemannian Optimization Perspective of the Gauss-Newton Method for Feedforward Neural Networks

2024-12-18 · Semih Cayci

We analyze the convergence of Gauss-Newton dynamics for training neural networks with smooth activation functions. In the underparameterized regime, the Gauss-Newton gradient flow induces a Riemannian gradient flow on a …

Riemannian optimization

Regularized Gauss-Newton for Optimizing Overparameterized Neural Networks

2024-04-23 · Adeyemi D. Adeoye, Philipp Christian Petersen, Alberto Bemporad

The generalized Gauss-Newton (GGN) optimization method incorporates curvature estimates into its solution steps, and provides a good approximation to the Newton method for large-scale optimization problems. GGN has been …

Gram-Gauss-Newton Method: Learning Overparameterized Neural Networks for Regression Problems

2019-05-28 · Tianle Cai, Ruiqi Gao, Jikai Hou, Siyu Chen 외

First-order methods such as stochastic gradient descent (SGD) are currently the standard algorithm for training deep neural networks. Second-order methods, despite their better convergence rate, are rarely used in practi…

regressionSecond-order methods

Error whitening: Why Gauss-Newton outperforms Newton

2026-05-11 · Maricela Best McKay, Nathan P. Lawrence, Brian Wetton, R. Bhushan Gopaluni arxiv

The Gauss-Newton matrix is widely viewed as a positive semidefinite approximation of the Hessian, yet mounting empirical evidence shows that Gauss-Newton descent outperforms Newton's method. We adopt a function space per…