FPGA implementation of back propagation neural network

FPGA implementation of back propagation neural network

This project presented a backpropagation neural network on FPGA which can conduct inference and training processes for linear and non-linear problems. The network structure chosen contains 3 input nodes, one hidden layer with three neuron units and 1 output node. In addition, this project compare...

Full description

Saved in:
Bibliographic Details
Main Author: Li, Jianing
Other Authors: Zheng Yuanjin
Format: Thesis-Master by Coursework
Language:English
Published: Nanyang Technological University 2022
Subjects:
Engineering::Electrical and electronic engineering::Applications of electronics
Online Access:https://hdl.handle.net/10356/159255
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:This project presented a backpropagation neural network on FPGA which can conduct inference and training processes for linear and non-linear problems. The network structure chosen contains 3 input nodes, one hidden layer with three neuron units and 1 output node. In addition, this project compares the training time between MATLAB and FPGA. The FPGA can achieve a much shorter training time owing to architecture advantage and computation data type simplification. In the end, the result of the neural network is displayed on the LEDs on the FPGA board. Keywords:

Similar Items