Method for Convolutional Neural Network Hardware Implementation Based on a Residue Number System

DOI10.1134/S0361768822080217MaRDI QIDQ6486103zbMATH OpenFDO

Authors M. V. Valueva, G. V. Valuev, Mikhail Babenko, Andrei Tchernykh, Jorge M. Cortés-Mendoza

Publication date 5 January 2023

Published in Programming and Computer Software (Search for Journal in Brave)

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Hardware implementations of nonnumerical algorithms (VLSI algorithms, etc.) (68W35)

Recommendations

• Application of the residue number system to reduce hardware costs of the convolutional neural network implementation
• Convolution accelerator designs using fast algorithms
• A faster algorithm for reducing the computational complexity of convolutional neural networks
• An FPGA implementation of deep spiking neural networks for low-power and fast classification
• Parallelization of cellular neural networks on GPU

Cited in

(7)

• Application of the residue number system to reduce hardware costs of the convolutional neural network implementation
• Convolution accelerator designs using fast algorithms
• A faster algorithm for reducing the computational complexity of convolutional neural networks
• An analytic formulation of convolutional neural network learning for pattern recognition
• FPGA design and hardware implementation of a convolutional neural network for classification of saccadic eye movements
• A neural network accelerated optimization method for FPGA
• Accelerating CNN models for face verification with convolution theorem

This page was built for publication: Method for Convolutional Neural Network Hardware Implementation Based on a Residue Number System

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6486103)