paper-with-me

Papers

Second-order Information in First-order Optimization Methods

2019-12-20 · Yuzheng Hu, Licong Lin, Shange Tang

In this paper, we try to uncover the second-order essence of several first-order optimization methods. For Nesterov Accelerated Gradient, we rigorously prove that the algorithm makes use of the difference between past and current gradients, thus approximates the Hessian and accelerates the training. For adaptive methods, we related Adam and Adagrad to a powerful technique in computation statistics---Natural Gradient Descent. These adaptive methods can in fact be treated as relaxations of NGD with only a slight difference lying in the square root of the denominator in the update rules. Skeptical about the effect of such difference, we design a new algorithm---AdaSqrt, which removes the square root in the denominator and scales the learning rate by sqrt(T). Surprisingly, our new algorithm is comparable to various first-order methods(such as SGD and Adam) on MNIST and even beats Adam on CIFAR-10! This phenomenon casts doubt on the convention view that the square root is crucial and training without it will lead to terrible performance. As far as we have concerned, so long as the algorithm tries to explore second or even higher information of the loss surface, then proper scaling of the learning rate alone will guarantee fast training and good generalization performance. To the best of our knowledge, this is the first paper that seriously considers the necessity of square root among all adaptive methods. We believe that our work can shed light on the importance of higher-order information and inspire the design of more powerful algorithms in the future.

📄 PDF Abstract BibTeX arXiv:1912.09926

Code (0)

등록된 구현이 없습니다.

Tasks

2D Human Pose Estimation

Methods 이 논문이 사용한 방법론

AdaSqrt 설명 없음
Nesterov Accelerated Gradient Nesterov Accelerated Gradient is a momentum-based SGD optimizer that "looks ahead" to where the parameters will be to calculate the…
AdaGrad AdaGrad is a stochastic optimization method that adapts the learning rate to the parameters. It performs smaller updates for parameters associated with frequently occurring…
Adam 설명 없음
SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…

Similar Papers 제목 키워드 기반

FOSI: Hybrid First and Second Order Optimization

2023-02-16 · Hadar Sivan, Moshe Gabel, Assaf Schuster

Popular machine learning approaches forgo second-order information due to the difficulty of computing curvature in high dimensions. We present FOSI, a novel meta-algorithm that improves the performance of any base first-…

Audio ClassificationLanguage ModellingSecond-order methodsTransfer Learning

Second-Order Stochastic Optimization for Machine Learning in Linear Time

2016-02-12 · Naman Agarwal, Brian Bullins, Elad Hazan

First-order stochastic methods are the state-of-the-art in large-scale machine learning optimization owing to efficient per-iteration complexity. Second-order methods, while able to provide faster convergence, have been …

BIG-bench Machine LearningSecond-order methodsStochastic Optimization

AdaFisher: Adaptive Second Order Optimization via Fisher Information

2024-05-26 · Damien Martins Gomes, Yanlei Zhang, Eugene Belilovsky, Guy Wolf 외

First-order optimization methods are currently the mainstream in training deep neural networks (DNNs). Optimizers like Adam incorporate limited curvature information by employing the diagonal matrix preconditioning of th…

Computational Efficiencyimage-classificationImage ClassificationLanguage Modeling+1

SGD with Partial Hessian for Deep Neural Networks Optimization

2024-03-05 · Ying Sun, Hongwei Yong, Lei Zhang

Due to the effectiveness of second-order algorithms in solving classical optimization problems, designing second-order optimizers to train deep neural networks (DNNs) has attracted much research interest in recent years.…

image-classificationImage ClassificationSecond-order methods

Towards Practical Second-Order Optimizers in Deep Learning: Insights from Fisher Information Analysis

2025-04-26 · Damien Martins Gomes

First-order optimization methods remain the standard for training deep neural networks (DNNs). Optimizers like Adam incorporate limited curvature information by preconditioning the stochastic gradient with a diagonal mat…

Computational Efficiencyimage-classificationImage ClassificationLanguage Modeling+2