Simple Neural Network can't learn XOR

Question

I\'m trying to learn about neural networks and coded a simple, back-propagation, neural network that uses sigmoid activation functions, random weight initialization, and learnin

Answer 1

My answer will be not about ruby, but about neural network. First of all, you have to understand how to write your inputs and your network on a paper. If you implement binary operatos, your space will consist of four points on XY-plane. Mark true and false on X and Y axis and draw your four points. If you to it right, you will receive something like this

Now(maybe you didn't know this interpretattion of neuron) try to draw neuron as a line on a plane, which separates your points as you need. For example, this is the line for AND: The line separates correct answers from incorrect. If you understand, you can write the line for OR. XOR will be a trouble.

And as a last step of this debug, realize a neuron as a line. Find a literature about it, I don't remember how to build neuron by existing line. It will be simple, really. Then build a neuron vector for AND implement it. Realize AND as a single neuron network, where neuron is defined as your AND, calculated on a paper. If you do all correct, your network will do AND function. I wrote such a huge number of letters just because you write a program before understanding a task. I don't want to be rough, but your mention of XOR showed it. If you will try to build XOR on one neuron, you will receive nothing - it's impossible to separate correct answers from incorrect. In books it is called "XOR is not linear separable". So for XOR you need to build a two layers network. For example, you will have AND and not-OR as a first layer and AND as a second layer.

If you still read this and you understand what I wrote, then you will have no troubles with debugging network. If your network fails to learn some function, then build it on a paper, then hardcode your network and test it. If It still fails, you build it on a paper incorrect - re-read my lecture;)