Benchmark Implementation

For this project, you can use any algorithm of your choosing to solve the task. You are strongly encouraged to do your own research, to devise your own approach towards solving this problem.

In case you get stuck, note that you should be able to solve the project by making only minor modifications to the DQN code provided as part of the Deep Q-Networks lesson. Please see the image below for an example of how you might expect your agent's score to evolve. If you're interested in about how long it should take, in the solution code for the project, we were able to solve the project in fewer than 1800 episodes.

Your agent might take longer (or solve the task much faster!) -- this is perfectly fine, and we provide this estimate only as an estimate for how long you should wait before assessing if your agent is learning.