A New Stochastic Computing Methodology for Efficient Neural Network Implementation

Canals Vincent; Morro Antoni; Oliver Antoni; Alomar Miquel L.; Rossello Josep L.

首页> 外文期刊>Neural Networks and Learning Systems, IEEE Transactions on >A New Stochastic Computing Methodology for Efficient Neural Network Implementation

【24h】

A New Stochastic Computing Methodology for Efficient Neural Network Implementation

机译：高效神经网络实现的一种新的随机计算方法

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

This paper presents a new methodology for the hardware implementation of neural networks (NNs) based on probabilistic laws. The proposed encoding scheme circumvents the limitations of classical stochastic computing (based on unipolar or bipolar encoding) extending the representation range to any real number using the ratio of two bipolar-encoded pulsed signals. Furthermore, the novel approach presents practically a total noise-immunity capability due to its specific codification. We introduce different designs for building the fundamental blocks needed to implement NNs. The validity of the present approach is demonstrated through a regression and a pattern recognition task. The low cost of the methodology in terms of hardware, along with its capacity to implement complex mathematical functions (such as the hyperbolic tangent), allows its use for building highly reliable systems and parallel computing.

机译：本文提出了一种基于概率定律的神经网络（NN）硬件实现的新方法。提出的编码方案克服了经典随机计算（基于单极性或双极性编码）的局限性，使用两个双极性编码脉冲信号的比率将表示范围扩展到任何实数。此外，由于其特定的编码方式，因此该新方法实际上具有总体的噪声免疫能力。我们介绍了用于构建实现NN所需的基本模块的不同设计。通过回归和模式识别任务证明了本方法的有效性。该方法在硬件方面的低成本以及其实现复杂数学功能（例如双曲线正切）的能力，使其可用于构建高度可靠的系统和并行计算。

著录项

来源
《Neural Networks and Learning Systems, IEEE Transactions on》 |2016年第3期|551-564|共14页
作者
Canals Vincent; Morro Antoni; Oliver Antoni; Alomar Miquel L.; Rossello Josep L.;
展开▼
作者单位

Department of Physics, University of Balearic Islands, Palma de Mallorca, Spain;

展开▼
收录信息
原文格式 PDF
正文语种 eng
中图分类
关键词
Computational intelligence; neural networks (NNs); pattern recognition; probabilistic logic; stochastic systems; stochastic systems.;

机译：计算智能;神经网络（NNs）;模式识别;概率逻辑;随机系统;随机系统。;

相似文献

外文文献
中文文献
专利

1. Efficient Implementation of Stochastic Computing Based Deep Neural Network on Low Cost Hardware with Saturation Arithmetic [J] . Sunny Bodiwala, Nirali Nanavati Journal of computer sciences . 2020,第11期

机译：用饱和算术高效实现低成本硬件的随机计算深神经网络
2. Efficient Implementation of Stochastic Computing Based Deep Neural Network on Low Cost Hardware with Saturation Arithmetic [J] . Sunny Bodiwala, Nirali Nanavati Journal of computer sciences . 2020,第11期

机译：高成本硬件与饱和算法的低成本硬件深神经网络的高效实现
3. Energy-efficient Design of MTJ-based Neural Networks with Stochastic Computing [J] . ACM Journal on Emerging Technologies in Computing Systems . 2020,第1期

机译：随机计算的基于MTJ的神经网络的节能设计
4. An Efficient Hardware Implementation of Activation Functions Using Stochastic Computing for Deep Neural Networks [C] . Van-Tinh Nguyen, Tieu-Khanh Luong, Han Le Duc, IEEE International Symposium on Embedded Multicore/Many-core System-on-Chip . 2018

机译：利用随机计算对深神经网络的有效硬件实现
5. A Design Methodology for Efficient Implementation of Deconvolutional Neural Networks on an FPGA [D] . Zhang, Xinyu. 2017

机译：在FPGA上高效实现反卷积神经网络的设计方法
6. ReStoCNet: Residual Stochastic Binary Convolutional Spiking Neural Network for Memory-Efficient Neuromorphic Computing [O] . Gopalakrishnan Srinivasan, Kaushik Roy 2010

机译：ReStoCNet：记忆有效的神经形态计算的残差随机二进制卷积穗状神经网络
7. Efficient Implementation of Stochastic Computing Based Deep Neural Network on Low Cost Hardware with Saturation Arithmetic [O] . Sunny Bodiwala, Nirali Nanavati 2020

机译：用饱和算术高效实现低成本硬件的随机计算深神经网络

A New Stochastic Computing Methodology for Efficient Neural Network Implementation

摘要

著录项

相似文献

相关主题

期刊订阅