arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.10044cs.NEcs.LG
收藏

BCNN:二元复合神经网络

BCNN: Binary Complex Neural Network

Zhejiang University(浙江大学);Pacific Northwest National Laboratory(太平洋西北国家实验室)

Yanfei Li, Tong Geng, Ang Li, Huimin Yu

摘要:

二值化神经网络(BNN)在资源有限的边缘侧应用中显示出巨大的前景,但引起了精度降低的担忧。受复杂神经网络的启发,本文将复表示引入BNN,提出了二进制复神经网络--一种通过复卷积处理二进制复输入和权值的新型网络设计,但仍然可以获得BNN非凡的计算效率。为了保证快速的收敛速度,我们提出了新的BCNN的批量归一化函数和权重初始化函数。使用最先进的网络模型(例如,ResNet,ResNetE和NIN)表明,与原始BNN模型相比,BCNN可以实现更好的准确性。BCNN通过复表示增强其学习能力并将其适用性扩展到复值输入数据来改进BNN。BCNN的源代码将在GitHub上发布。

英文摘要:

Binarized neural networks, or BNNs, show great promise in edge-side applications with resource limited hardware, but raise the concerns of reduced accuracy. Motivated by the complex neural networks, in this paper we introduce complex representation into the BNNs and propose Binary complex neural network -- a novel network design that processes binary complex inputs and weights through complex convolution, but still can harvest the extraordinary computation efficiency of BNNs. To ensure fast convergence rate, we propose novel BCNN based batch normalization function and weight initialization function. Experimental results on Cifar10 and ImageNet using state-of-the-art network models (e.g., ResNet, ResNetE and NIN) show that BCNN can achieve better accuracy compared to the original BNN models. BCNN improves BNN by strengthening its learning capability through complex representation and extending its applicability to complex-valued input data. The source code of BCNN will be released on GitHub.

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。

↑