On the Challenge of Training Small Scale Neural Networks on Large Scale Computing Systems

被引：0

作者：

Malysiak, Darius ^{[1
]}

Grimm, Matthias ^{[1
]}

机构：

[1] Hsch Ruhr West, Inst Comp Sci, Bottrop, Germany

来源：

2015 16TH IEEE INTERNATIONAL SYMPOSIUM ON COMPUTATIONAL INTELLIGENCE AND INFORMATICS (CINTI) | 2015年

关键词：

statistics; gradient descent; gpgpu; high performance computing; neural networks; backpropagation; opencl; cuda;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present a novel approach of distributing small-to mid-scale neural networks onto modern parallel architectures. In this context we discuss the induced challenges and possible solutions. We provide a detailed theoretical analysis with respect to space and time complexities and reinforce our computation model with evaluations which show a performance gain over state of the art approaches.

引用

页码：273 / 284

页数：12