arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1809.09244cs.LGstat.ML
收藏

没有乘法?没有浮点数?没问题!训练网络以实现高效推理

No Multiplication? No Floating Point? No Problem! Training Networks for Efficient Inference

Google Research(谷歌研究院)

Shumeet Baluja, David Marwood, Michele Covell, Nick Johnston

摘要:

为了在高度资源受限的设备(助听器、耳塞、可穿戴设备)上成功部署深度神经网络,我们必须简化运算类型和推理过程中使用的内存/电源资源。完全避免推理时浮点运算是为这些高度受限的环境设计网络的最简单方法之一。通过对网络内的非线性和网络权重进行离散化,我们可以转向简单、紧凑的网络,无需浮点运算,无需乘法,并避免所有非线性函数计算。我们的方法使我们能够探索可能的网络的频谱,从完全连续的版本到具有双层权重和激活的网络。我们的研究结果表明,离散化可以在不损失性能的情况下完成,并且我们可以训练一个网络,该网络将在没有浮点数,没有乘法,并且在回归任务(自动编码)和多类分类任务(ImageNet)上使用更少的RAM成功运行。部署我们的离散化网络所需的内存不到使用浮点运算的等效架构的三分之一。

英文摘要:

For successful deployment of deep neural networks on highly--resource-constrained devices (hearing aids, earbuds, wearables), we must simplify the types of operations and the memory/power resources used during inference. Completely avoiding inference-time floating-point operations is one of the simplest ways to design networks for these highly-constrained environments. By discretizing both our in-network non-linearities and our network weights, we can move to simple, compact networks without floating point operations, without multiplications, and avoid all non-linear function computations. Our approach allows us to explore the spectrum of possible networks, ranging from fully continuous versions down to networks with bi-level weights and activations. Our results show that discretization can be done without loss of performance and that we can train a network that will successfully operate without floating-point, without multiplication, and with less RAM on both regression tasks (auto encoding) and multi-class classification tasks (ImageNet). The memory needed to deploy our discretized networks is less than one third of the equivalent architecture that does use floating-point operations.

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。

↑