Abstract
Ant Colony Optimization is a meta-heuristic approach to solve difficult optimization problems. Training a neural network is a process of finding the optimal set of its connection weights. So, a Continuous Ant Colony Optimization algorithm is used to train the neural network. In this project, the continuous Ant Colony Optimization (ACO) algorithm used to train neural networks was studied, implemented and tested with known training problems. Finally, the performance of this ACO implementation was compared with that of backpropagation and found to be less effective than backpropagation.