arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1708.04728cs.CV
收藏

DeepRebirth:加速移动设备上的深度神经网络执行

DeepRebirth: Accelerating Deep Neural Network Execution on Mobile Devices

Samsung Research America(三星美国研究院)

Dawei Li, Xiaolong Wang, Deguang Kong

摘要:

在移动设备上部署深度神经网络是一项具有挑战性的任务。现有的矩阵分解等模型压缩方法有效地减小了部署模型的规模,但仍不能满足实时处理的要求。本文首先发现,主要的障碍是非张量层(如池化和归一化)的执行时间过长,而没有类似张量的可训练参数。这促使我们设计一个新的加速框架:通过“瘦身”现有的连续和平行的非张量和张量层的DeepRebirth。层瘦身在不同的子结构上执行:(a)通过垂直合并连续的非张量层和张量层进行流线瘦身;(b)通过水平合并非张量分支和张量分支进行分支瘦身。所提出的优化操作显着加快了模型的执行,也大大降低了运行时的内存成本,因为精简的模型架构包含更少的隐藏层。为了最大限度地避免精度损失,在新生成的层中的参数学习与逐层微调的基础上的理论分析和经验验证。正如实验中所观察到的,DeepRebirth在GoogLeNet上实现了超过3倍的加速和2.5倍的运行时内存节省,而ImageNet上的前5名准确率仅下降了0.4%。此外,通过与其他模型压缩技术相结合,DeepRebirth在三星Galaxy S6的CPU上提供了平均65 ms的推理时间,前5名的准确率为86.5%,比SqueezeNet快14%,后者的前5名准确率仅为80.5%。

英文摘要:

Deploying deep neural networks on mobile devices is a challenging task. Current model compression methods such as matrix decomposition effectively reduce the deployed model size, but still cannot satisfy real-time processing requirement. This paper first discovers that the major obstacle is the excessive execution time of non-tensor layers such as pooling and normalization without tensor-like trainable parameters. This motivates us to design a novel acceleration framework: DeepRebirth through "slimming" existing consecutive and parallel non-tensor and tensor layers. The layer slimming is executed at different substructures: (a) streamline slimming by merging the consecutive non-tensor and tensor layer vertically; (b) branch slimming by merging non-tensor and tensor branches horizontally. The proposed optimization operations significantly accelerate the model execution and also greatly reduce the run-time memory cost since the slimmed model architecture contains less hidden layers. To maximally avoid accuracy loss, the parameters in new generated layers are learned with layer-wise fine-tuning based on both theoretical analysis and empirical verification. As observed in the experiment, DeepRebirth achieves more than 3x speed-up and 2.5x run-time memory saving on GoogLeNet with only 0.4% drop of top-5 accuracy on ImageNet. Furthermore, by combining with other model compression techniques, DeepRebirth offers an average of 65ms inference time on the CPU of Samsung Galaxy S6 with 86.5% top-5 accuracy, 14% faster than SqueezeNet which only has a top-5 accuracy of 80.5%.

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。

↑