paper-with-me

홈 › Papers

Layer-Parallel Training of Residual Networks with Auxiliary Variables

2021-09-27 · NeurIPS Workshop DLDE 2021 12 · Qi Sun, Hexin Dong, Zewei Chen, Weizhen Dian, Jiacheng Sun, Yitong Sun, Zhenguo Li, Bin Dong

Backpropagation algorithm is indispensable for training modern residual networks (ResNets) and usually tends to be time-consuming due to its inherent algorithmic lockings. Auxiliary-variable methods, e.g., the penalty and augmented Lagrangian (AL) methods, have attracted much interest lately due to their ability to exploit layer5 wise parallelism. However, we find that large communication overhead and lacking data augmentation are two key challenges of these approaches, which may lead to low speedup and accuracy drop. Inspired by the continuous-time formulation of ResNets, we propose a novel serial-parallel hybrid (SPH) training strategy to enable the use of data augmentation during training, together with downsampling (DS) filters to reduce the communication cost. This strategy first trains the network by solving a succession of independent sub-problems in parallel and then improve the trained network through a full serial forward-backward propagation of data. We validate our methods on modern ResNets across benchmark datasets, achieving speedup over the backpropagation while maintaining comparable accuracy.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Data Augmentation

Similar Papers 제목 키워드 기반

Layer-Parallel Training of Residual Networks with Auxiliary-Variable Networks

2021-12-10 · Qi Sun, Hexin Dong, Zewei Chen, Jiacheng Sun 외

Gradient-based methods for the distributed training of residual networks (ResNets) typically require a forward pass of the input data, followed by back-propagating the error gradient to update model parameters, which bec…

Data Augmentation

A Practical Layer-Parallel Training Algorithm for Residual Networks

2020-09-03 · Qi Sun, Hexin Dong, Zewei Chen, Weizhen Dian 외

Gradient-based algorithms for training ResNets typically require a forward pass of the input data, followed by back-propagating the objective gradient to update parameters, which are time-consuming for deep ResNets. To b…

Data Augmentation

Layer-Parallel Training of Deep Residual Neural Networks

2018-12-11 · S. Günther, L. Ruthotto, J. B. Schroder, E. C. Cyr 외

Residual neural networks (ResNets) are a promising class of deep neural networks that have shown excellent performance for a number of learning tasks, e.g., image classification and recognition. Mathematically, ResNet ar…

General Classificationimage-classificationImage Classification

Layer-Parallel Training with GPU Concurrency of Deep Residual Neural Networks via Nonlinear Multigrid

2020-07-14 · Andrew C. Kirby, Siddharth Samsi, Michael Jones, Albert Reuther 외

A Multigrid Full Approximation Storage algorithm for solving Deep Residual Networks is developed to enable neural network parallelized layer-wise training and concurrent computational kernel execution on GPUs. This work …

GPU

Residual CNDS

2016-08-07 · Hussein A. Al-Barazanchi, Hussam Qassim, Abhishek Verma

Convolutional Neural networks nowadays are of tremendous importance for any image classification system. One of the most investigated methods to increase the accuracy of CNN is by increasing the depth of CNN. Increasing …

General Classificationimage-classificationImage Classification