Reading
Neural Networks
Eric Roberts
This web-based tutorial, based at Stanford University, discusses several topics pertaining to ANNs, including their biological inspiration, history, ANN architecture, ANN applications, and other useful material. Some of the interactive features are no longer supported on current browsers, but the static (still) images and graphics are very informative. The material on this website can be profitably studied for a basic, yet thorough introduction to the fundamental concepts of artificial neural networks.
A Step-by-Step Neural Network Tutorial for Beginners
Tirmidzi Faizal Aflahi
Jun 19, 2019
This web article provides a non-mathematical introduction to artificial neural networks. It addresses the place of ANNs in machine learning, applications, training, validation, and testing, a description of the MNIST data set, and a basic Python implementation. However, the Python examples discuss the Keras machine learning implementation, whereas the examples in this section use Scikit Learn. However, the user will benefit from the presentation of the basic concepts.
Learning and Neural Networks
This website provides additional explanations, examples, and applications of ANNs. Some of the material on this website is mathematically complex. However, the article may still be profitably read by “glossing over” some of the more advanced mathematical concepts, and study the examples and illustrations.
Neural Network Models (Supervised)
This Scikit Learn website contains a discussion of basic ANN principles, basing the development on the fundamentals of perceptrons. Constructing ANN models in Scikit Learn is presented.
Visualization of MLP weights on MNIST
This Scikit Learn page explains how the weights (coefficients) that an ANN has learned can be visualized for analyzing network learning behaviour.
MNIST Data
This web site contains images of handwritten characters from the MNIST database.
Python Code
This section uses the Python code ANN_MNIST_Example.py and the MNIST data files.