General Advice

In deep RL, failure comes in many flavors. Sometimes your agent doesn't train at all. Sometimes it appears to be learning, but then something weird happens late in training that completely derails its performance.

When this happens, remember that you are part of a community. Please reach out to your fellow students in Slack, ask for advice in Knowledge, submit your project to get feedback from a reviewer, or ask your instructors in office hours. We will figure this out together! Deep RL isn't easy and arguably "doesn't work yet". A large part of mastering cutting-edge algorithms is banging your head against the table - and we'll be right there with you. We promise that it will be so incredibly rewarding when you ultimately get it working!

## Failure is Inevitable and Only Temporary

When we wrote a solution to this project, we didn't get it right the first time. Or the second. Or the third. :)