A simple demonstration of reinforcement learning using a Neural Network model to play the Cartpole game .
Reinforcement learning is implemented wherein a Neural Network Model learns to play the Cartpole game and makes decisions regarding the next move depending upon the information of the current game scenario it got from the previous move .The model was initially trained on random moves and then trained on calculated risk moves .
An average score of 188 is expected over 100 turns of the game . The following model got an accuracy of 200 points for 15 turns .