paper-with-me

Papers

Accelerating SGD for Distributed Deep-Learning Using Approximated Hessian Matrix

2017-09-15 · Sébastien M. R. Arnold, Chunming Wang

We introduce a novel method to compute a rank $m$ approximation of the inverse of the Hessian matrix in the distributed regime. By leveraging the differences in gradients and parameters of multiple Workers, we are able to efficiently implement a distributed approximation of the Newton-Raphson method. We also present preliminary results which underline advantages and challenges of second-order methods for large stochastic optimization problems. In particular, our work suggests that novel strategies for combining gradients provide further information on the loss surface.

📄 PDF Abstract BibTeX arXiv:1709.05069

Code (0)

등록된 구현이 없습니다.

Tasks

Deep LearningSecond-order methodsStochastic Optimization

Similar Papers 제목 키워드 기반

Accelerating Sinkhorn Algorithm with Sparse Newton Iterations

2024-01-20 · Xun Tang, Michael Shavlovsky, Holakou Rahmanian, Elisa Tardini 외

Computing the optimal transport distance between statistical distributions is a fundamental task in machine learning. One remarkable recent advancement is entropic regularization and the Sinkhorn algorithm, which utilize…

FAGH: Accelerating Federated Learning with Approximated Global Hessian

2024-03-16 · Mrinmay Sen, A. K. Qin, Krishna Mohan C

In federated learning (FL), the significant communication overhead due to the slow convergence speed of training the global model poses a great challenge. Specifically, a large number of communication rounds are required…

Federated Learning

Kronecker-factored Quasi-Newton Methods for Deep Learning

2021-02-12 · Yi Ren, Achraf Bahamou, Donald Goldfarb

Second-order methods have the capability of accelerating optimization by using much richer curvature information than first-order methods. However, most are impractical for deep learning, where the number of training par…

Deep LearningSecond-order methods

Newton-Type Methods for Non-Convex Optimization Under Inexact Hessian Information

2017-08-23 · Peng Xu, Fred Roosta, Michael W. Mahoney

We consider variants of trust-region and cubic regularization methods for non-convex optimization, in which the Hessian matrix is approximated. Under mild conditions on the inexact Hessian, and using approximate solution…

Vocal Bursts Type Prediction

Distributed estimation of the inverse Hessian by determinantal averaging

2019-05-28 · NeurIPS 2019 12 · Michał Dereziński, Michael W. Mahoney

In distributed optimization and distributed numerical linear algebra, we often encounter an inversion bias: if we want to compute a quantity that depends on the inverse of a sum of distributed matrices, then the sum of t…

Distributed OptimizationUncertainty Quantification