English
Related papers

Related papers: A Trace-restricted Kronecker-Factored Approximatio…

200 papers

Distributed training with synchronous stochastic gradient descent (SGD) on GPU clusters has been widely used to accelerate the training process of deep models. However, SGD only utilizes the first-order gradient in model parameter updates,…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-07-15 Shaohuai Shi , Lin Zhang , Bo Li

A key challenge for gradient based optimization methods in model-free reinforcement learning is to develop an approach that is sample efficient and has low variance. In this work, we apply Kronecker-factored curvature estimation technique…

Machine Learning · Computer Science 2018-12-12 Mohammad Firouzi

In this work, we propose to apply trust region optimization to deep reinforcement learning using a recently proposed Kronecker-factored approximation to the curvature. We extend the framework of natural policy gradient and propose to…

Machine Learning · Computer Science 2017-08-21 Yuhuai Wu , Elman Mansimov , Shun Liao , Roger Grosse , Jimmy Ba

Despite the predominant use of first-order methods for training deep learning models, second-order methods, and in particular, natural gradient methods, remain of interest because of their potential for accelerating training through the use…

Machine Learning · Computer Science 2021-12-23 Yi Ren , Donald Goldfarb

The Variational Monte Carlo method has recently seen important advances through the use of neural network quantum states. While more and more sophisticated ans\"atze have been designed to tackle a wide variety of quantum many-body problems,…

Nuclear Theory · Physics 2025-07-09 M. Drissi , J. W. T. Keeble , J. Rozalén Sarmiento , A. Rios

Recently, optimizers that explicitly treat weights as matrices, rather than flattened vectors, have demonstrated their effectiveness. This perspective naturally leads to structured approximations of the Fisher matrix as preconditioners,…

Machine Learning · Computer Science 2025-11-11 Nikolay Yudin , Ekaterina Grishina , Andrey Veprikov , Alexandr Beznosikov , Maxim Rakhuba

K-FAC is a successful tractable implementation of Natural Gradient for Deep Learning, which nevertheless suffers from the requirement to compute the inverse of the Kronecker factors (through an eigen-decomposition). This can be very…

Machine Learning · Computer Science 2022-11-28 Constantin Octavian Puiu

We propose a quadratic penalty method for continual learning of neural networks that contain batch normalization (BN) layers. The Hessian of a loss function represents the curvature of the quadratic penalty function, and a…

Machine Learning · Computer Science 2020-04-17 Janghyeon Lee , Hyeong Gwon Hong , Donggyu Joo , Junmo Kim

Online Continual Learning (OCL) for image classification represents a challenging subset of Continual Learning, focusing on classifying images from a stream without assuming data independence and identical distribution (i.i.d). The primary…

Machine Learning · Computer Science 2026-03-24 Joe Khawand , David Colliaux

In this technical report, we consider an approach that combines the PPO objective and K-FAC natural gradient optimization, for which we call PPOKFAC. We perform a range of empirical analysis on various aspects of the algorithm, such as…

Artificial Intelligence · Computer Science 2018-01-18 Jiaming Song , Yuhuai Wu

K-FAC (arXiv:1503.05671, arXiv:1602.01407) is a tractable implementation of Natural Gradient (NG) for Deep Learning (DL), whose bottleneck is computing the inverses of the so-called ``Kronecker-Factors'' (K-factors). RS-KFAC…

Machine Learning · Computer Science 2023-09-13 Constantin Octavian Puiu

Modern GPUs are equipped with large amounts of high-bandwidth memory, enabling them to support mini-batch sizes of up to tens of thousands of training samples. However, most existing optimizers struggle to perform effectively at such a…

Machine Learning · Computer Science 2026-02-10 Yishun Lu , Wesley Armour

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 matrix. Despite the…

Machine Learning · Computer Science 2025-04-30 Damien Martins Gomes

Despite all the impressive advances of recurrent neural networks, sequential data is still in need of better modelling. Truncated backpropagation through time (TBPTT), the learning algorithm most widely used in practice, suffers from the…

Machine Learning · Computer Science 2018-12-07 Asier Mujika , Florian Meier , Angelika Steger

Second-order optimization algorithms exhibit excellent convergence properties for training deep learning models, but often incur significant computation and memory overheads. This can result in lower training efficiency than the first-order…

Machine Learning · Computer Science 2023-08-07 Lin Zhang , Shaohuai Shi , Bo Li

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 the stochastic…

Machine Learning · Computer Science 2025-03-12 Damien Martins Gomes , Yanlei Zhang , Eugene Belilovsky , Guy Wolf , Mahdi S. Hosseini

This paper introduces a new stochastic optimization method based on the regularized Fisher information matrix (FIM), named SOFIM, which can efficiently utilize the FIM to approximate the Hessian matrix for finding Newton's gradient update…

Machine Learning · Computer Science 2024-05-02 Mrinmay Sen , A. K. Qin , Gayathri C , Raghu Kishore N , Yen-Wei Chen , Balasubramanian Raman

Large-scale distributed training of deep neural networks suffer from the generalization gap caused by the increase in the effective mini-batch size. Previous approaches try to solve this problem by varying the learning rate and batch size…

Machine Learning · Computer Science 2019-04-02 Kazuki Osawa , Yohei Tsuji , Yuichiro Ueno , Akira Naruse , Rio Yokota , Satoshi Matsuoka

Differentially private federated learning (DP-FL) often suffers from slow convergence under tight privacy budgets because the noise required for privacy preservation degrades gradient quality. Although second-order optimization can…

Machine Learning · Computer Science 2026-03-25 Sidhant Nair , Tanmay Sen , Mrinmay Sen , Sayantan Banerjee

Research into optimisation for deep learning is characterised by a tension between the computational efficiency of first-order, gradient-based methods (such as SGD and Adam) and the theoretical efficiency of second-order, curvature-based…

Machine Learning · Computer Science 2024-06-17 Ross M. Clarke , José Miguel Hernández-Lobato