Faculty Advisor

Gennert, Michael A.

Abstract

This Major Qualifying Project is an open source supervised machine learning neural network software development kit (SDK) utilizing a stochastic gradient descent with cross-entropy error backpropagation algorithm, implemented in Java. This SDK provides a human-readable implementation of a commonly used method of machine learning, thereby supplying a transparent learning tool for students. The SKD was demonstrated by providing functional examples of networks designed for classification. Examples were provided ranging in complexity from the simplest most transparent example of learning, to a real world example of classification based on color.

Publisher

Worcester Polytechnic Institute

Date Accepted

April 2017

Major

Computer Science

Project Type

Major Qualifying Project

Accessibility

Unrestricted

Advisor Department

Computer Science

Share

COinS