arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2009.01462cs.LGcs.AIstat.ML
收藏

一种实用的残差网络分层并行训练算法

A Practical Layer-Parallel Training Algorithm for Residual Networks

Peking University(北京大学);Huawei Noah’s Ark Lab(华为诺亚方舟实验室)

Qi Sun, Hexin Dong, Zewei Chen, Weizhen Dian, Jiacheng Sun, Yitong Sun, Zhenguo Li, Bin Dong

摘要:

用于训练ResNet的基于递归的算法通常需要前向传递输入数据,然后反向传播目标梯度以更新参数,这对于深度ResNet来说是耗时的。为了打破模块之间的依赖关系,在向前和向后的模式,罚函数和增广拉格朗日(AL)的方法,最近引起了很大的兴趣,由于他们的能力,利用逐层并行。然而,我们观察到,大的通信开销和缺乏数据增强是这些方法的两个关键挑战,这可能导致低加速比和跨多个计算设备的准确性下降。受ResNets最优控制公式的启发,我们提出了一种新的串并行混合训练策略,以使用数据增强,以及下采样滤波器来降低通信成本。该策略首先通过并行求解一系列独立的子问题来训练网络参数,然后通过数据的全串行前向-后向传播来校正网络参数。这种策略可以应用于大多数现有的使用辅助变量的层并行训练方法。作为一个例子,我们在MNIST、CIFAR-10和CIFAR-100数据集上使用ResNet和WideResNet上的惩罚和AL方法验证了所提出的策略,与传统的层-序列训练方法相比,实现了显著的加速,同时保持了相当的准确性。

英文摘要:

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 break the dependencies between modules in both the forward and backward modes, auxiliary-variable methods such as the penalty and augmented Lagrangian (AL) approaches have attracted much interest lately due to their ability to exploit layer-wise parallelism. However, we observe that large communication overhead and lacking data augmentation are two key challenges of these methods, which may lead to low speedup ratio and accuracy drop across multiple compute devices. Inspired by the optimal control formulation of ResNets, we propose a novel serial-parallel hybrid training strategy to enable the use of data augmentation, together with downsampling filters to reduce the communication cost. The proposed strategy first trains the network parameters by solving a succession of independent sub-problems in parallel and then corrects the network parameters through a full serial forward-backward propagation of data. Such a strategy can be applied to most of the existing layer-parallel training methods using auxiliary variables. As an example, we validate the proposed strategy using penalty and AL methods on ResNet and WideResNet across MNIST, CIFAR-10 and CIFAR-100 datasets, achieving significant speedup over the traditional layer-serial training methods while maintaining comparable accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑