ECE 535 Artificial Neural Networks

Homework 2 Due: 28 April 2016 In this homework, you are going to design and test a 4-neuron Hopfield network. The activation

function is the signum function. The network should store the following patterns: y1 =        

1 1 1 1         y2 =

        −1 −1 1 1

        y3 =        

−1 1 − 1 1

        Note that these patterns are mutually orthogonal.

1. Find the weight matrix using the outer product rule. 2. Implement the network using MATLAB for both synchronous update and asynchronous update schemes. In the asynchronous update implementation, randomize the order of neurons that are being updated.

3. For all 16 possible initial states, run the network and find which pattern each initial state converges to and in how many steps. Use both synchronous and asynchronous update schemes.