Learning XOR Function with 2-layer Neural Network

A bare bones implementation of neural network with back-prop for evaluation of gradient. Minimization performed with a gradient descent.

Training Data with label

(0,0) --> 0

(0,1) --> 1

(1,0) --> 1

(1,1) --> 0

learn_xor_hidden_bprp.m - Main Script. **Run this file

forward_pass.m - Function to evaluate the forward pass

forward_pass_predict.m - Function to just make prediction with forward pass

backward_pass.m - Evaluation of gradient with back propagation

eval_performance.m - evaluation of the performance of the trained network