Full Source: https://github.com/royce-mathew/ForexPredictor/blob/master/src/ForexPredictor.ipynb
Predicting the forex market can be quite hard for a neural network to understand due to the properties of the market. To perfectly predict the market, the neural network would need a very high number of parameters and thousands of iterations of training.
This paper tries to explore the question on whether a neural network is able to predict the market given a limited set of parameters.
The program works by using neural networking algorithm known as NEAT. "NEAT" stands for "Neural Networks through Augmented Topologies". The neat algorithm works by using the concept of life, where each generation starts to learn based on the last generations, and the best of each generation gets to pass onto the next generation.
A proper explanation of how the algorithm works can be found on https://www.cs.ucf.edu/~kstanley/neat.html.