I have made a simple MLP which tries to play the cart-pole game. The weights of the neural network is derived using simple genetic algorithm.
I have included the optimals weights and the architecture in the test file in order to run the simulation with the derived weights.