paper-with-me

홈 › Papers

Parallel training of DNNs with Natural Gradient and Parameter Averaging

2014-10-27 · Daniel Povey, Xiaohui Zhang, Sanjeev Khudanpur

We describe the neural-network training framework used in the Kaldi speech recognition toolkit, which is geared towards training DNNs with large amounts of training data using multiple GPU-equipped or multi-core machines. In order to be as hardware-agnostic as possible, we needed a way to use multiple machines without generating excessive network traffic. Our method is to average the neural network parameters periodically (typically every minute or two), and redistribute the averaged parameters to the machines for further training. Each machine sees different data. By itself, this method does not work very well. However, we have another method, an approximate and efficient implementation of Natural Gradient for Stochastic Gradient Descent (NG-SGD), which seems to allow our periodic-averaging method to work well, as well as substantially improving the convergence of SGD on a single machine.

📄 PDF Abstract BibTeX arXiv:1410.7455

Code (1)

YiwenShaoStephen/NGD-SGD pytorch

Tasks

GPUspeech-recognitionSpeech Recognition

Methods 이 논문이 사용한 방법론

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

Understanding Why Neural Networks Generalize Well Through GSNR of Parameters

2020-01-21 · ICLR 2020 1 · Jinlong Liu, Guoqing Jiang, Yunzhi Bai, Ting Chen 외

As deep neural networks (DNNs) achieve tremendous success across many application domains, researchers tried to explore in many aspects on why they generalize well. In this paper, we provide a novel perspective on these …

Data-Parallel Neural Network Training via Nonlinearly Preconditioned Trust-Region Method

2025-02-07 · Samuel A. Cruz Alegría, Ken Trotti, Alena Kopaničáková, Rolf Krause

Parallel training methods are increasingly relevant in machine learning (ML) due to the continuing growth in model and dataset sizes. We propose a variant of the Additively Preconditioned Trust-Region Strategy (APTS) for…

Empirical Evaluation of Parallel Training Algorithms on Acoustic Modeling

2017-03-17 · Wenpeng Li, Bin-Bin Zhang, Lei Xie, Dong Yu

Deep learning models (DLMs) are state-of-the-art techniques in speech recognition. However, training good DLMs can be time consuming especially for production-size models and corpora. Although several parallel training a…

GPUspeech-recognitionSpeech Recognition

Quantized Distributed Training of Large Models with Convergence Guarantees

2023-02-05 · Ilia Markov, Adrian Vladu, Qi Guo, Dan Alistarh

Communication-reduction techniques are a popular way to improve scalability in data-parallel training of deep neural networks (DNNs). The recent emergence of large language models such as GPT has created the need for new…

Quantization

Hierarchical Weight Averaging for Deep Neural Networks

2023-04-23 · Xiaozhe Gu, Zixun Zhang, Yuncheng Jiang, Tao Luo 외

Despite the simplicity, stochastic gradient descent (SGD)-like algorithms are successful in training deep neural networks (DNNs). Among various attempts to improve SGD, weight averaging (WA), which averages the weights o…