Computational Analysis of stochastic arithmetic computing and stochastic activation function

Abstract

Stochastic Number Generators (SNG) plays a significant role in designing a stochastic computing system. SNGs make the stochastic system comfortable for computing in the stochastic domain. The challenges in developing the stochastic computing system are correlation and hardware area occupancy. By considering these phenomena, we have considered Linear Feedback Shift Register (LFSR) based SNG and S-box based SNG in this work. Our contributions to this paper are stochastic computation for activation functions using the SNGs mentioned above and stochastic computation for arithmetic components in the stochastic domain. By considering the two SNG methods, the difference in the computation for accuracy has been analyzed for stochastic activation function and stochastic arithmetic computation. The better SNG method will be used as SNG for stochastic convolutional neural network design using this analysis.

Accuracy Analysis on Design of Stochastic Computing in Arithmetic Components and Combinational Circuit

“…An approach for binarized neural networks was developed for the Modified National Institute of Standards and Technology database(MNIST) and Canadian Institute For Advanced Research(CIFAR) data focusing on performance of the convolutional binarized network. This has been extended for ASIC implementations [29][30][31][32]. The authors have suggested the importance of a CNN accelerator and the use of Python as an Algorithmic High-Level Language, for the realization and the implementation and FPGA based accelerators for CNNs.…”

“…The main focus of our research work is to build an efficient SC system focusing on accuracy improvement at all levels till the development of the CNN system. As a first step towards the same an S-Box-based the SNG methodology, (where-in the S-Box [34,35] in cryptographic algorithm) is used as the stochastic bit stream generator for building up the SC components. The developed components will be used for the development of CNN layers.…”