I did an Q-learning algorithm that work well, so I would try to do a DeepQ-learning algorithm to learn the CartPole env of gym env. So based on my Q-learning algorith I tried to evolve it on a DeepQ-learning algorithm without sucess during 2-3 days.
My problem is I just don’t understand why this is not working :
-Is it because I did a dumb mistake ?
-Is it because my model don’t converge because I update it every step ?
-Is it because I don’t use the right parameters ?