value functions. In those approaches, instead of learning the optimal action in a given state, the agent learns to predict how good a given state or action will be for the agent to be in.

Simple Reinforcement Learning in Tensorflow: Part 1 - Two-armed Bandit

Arthur Juliani

2.4K23

Hi Arthur, my name is Nitin Kanwar. I am a graduate student doing research in the field of reinforcement learning. I was wondering if you could provide me with a path in better understanding this field. By path I mean a few pointers on the things which I should revise/know before I dive head first. I know, I need to learn topics like probability, linear algebra etc, but is there anything else that I should focus on. Also, in order to approach Deep Reinforcement Learning, what all other technologies should I be familiar with. Like Neural Networks, Deep Learning etc. Thank you for the amazing article and looking forward to hearing from you.