arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.09052cs.LGcs.CVstat.ML
收藏

宽压缩:张量环网

Wide Compression: Tensor Ring Nets

Purdue University(普渡大学);Technicolor Research(泰克尼科尔研究(Technicolor Research));Adobe(Adobe公司);Duke University(杜克大学)

Wenqi Wang, Yifan Sun, Brian Eriksson, Wenlin Wang, Vaneet Aggarwal

摘要:

深度神经网络已经在各种现实世界的应用中展示了最先进的性能。为了获得性能增益,这些网络变得越来越大,越来越深,包含数百万甚至数十亿个参数和上千层。代价是这些大型架构需要大量的内存、存储和计算,从而限制了它们的可用性。受最近的张量环分解的启发,我们引入了张量环网络(TR-Nets),它显著压缩了深度神经网络的全连接层和卷积层。我们的研究结果表明,我们的TR-网络方法{能够通过$11\times$压缩LeNet-5而不损失准确性},并且可以通过$243\times$压缩最先进的Wide ResNet,在{Cifar 10图像分类}中只有2.3\%的退化。总的来说,这种压缩方案在科学计算和深度学习中显示出了前景,特别是对于新兴的资源受限设备,如智能手机,可穿戴设备和物联网设备。

英文摘要:

Deep neural networks have demonstrated state-of-the-art performance in a variety of real-world applications. In order to obtain performance gains, these networks have grown larger and deeper, containing millions or even billions of parameters and over a thousand layers. The trade-off is that these large architectures require an enormous amount of memory, storage, and computation, thus limiting their usability. Inspired by the recent tensor ring factorization, we introduce Tensor Ring Networks (TR-Nets), which significantly compress both the fully connected layers and the convolutional layers of deep neural networks. Our results show that our TR-Nets approach {is able to compress LeNet-5 by $11\times$ without losing accuracy}, and can compress the state-of-the-art Wide ResNet by $243\times$ with only 2.3\% degradation in {Cifar10 image classification}. Overall, this compression scheme shows promise in scientific computing and deep learning, especially for emerging resource-constrained devices such as smartphones, wearables, and IoT devices.

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。

↑