Update: For a newer neural net simulator optimized for image processing, see neural2d.net.
Update: For a beginner's introduction to the concepts and abstractions needed to understand how neural nets learn and work, and for tips for preparing training data for your neural net, see the new companion video "The Care and Training of Your Backpropagation Neural Net" at vimeo.com/technotes/neural-net-care-and-training .
Neural nets are fun to play with. Join me as we design and code a classic back-propagation neural net in C++, with adjustable gradient descent learning and adjustable momentum. Then train your net to do amazing and wonderful things. More at the blog: millermattson.com/dave/
For the neural net beginner, an introduction to all the concepts and abstractions you need to know in order to gain an intuitive understanding of how these crafty little neural nets learn to do anything at all. We'll discuss practical tips on preparing training data, and strategies for solving various kinds of problems. We'll discuss:
1. Review of how a backprop NN works
2. Type and range of the input and output data
3. Preparing training data
4. Training strategies
5. Adjustable parameters for learning rate and momentum
6. Visualizing how a NN solves a problem - a geometric interpretation
Best viewed in HD 720p.
(Sorry, this is not a programming tutorial; for that, please see the companion video, "Make a Neural Net Simulator in C++" at vimeo.com/19569529 .)