DeepTraffic is a deep reinforcement learning competition hosted as part of the MIT Deep Learning courses. The goal is to create a neural network that drives a vehicle (or multiple vehicles) as fast as possible through dense highway traffic. Top 10 submissions are listed on the leaderboard and you'll be able to visualize your submission in the following way: To get started right away, this repository provides a code snippet to insert into the code box on the DeepTraffic site. We'll add additional agents as the course progresses: A basic network that achieves 66.8mph. And now let's return to the problem of traffic: "Americans will put up with anything provided it doesn't block traffic." - Dan Rather In the U.S. alone, we spend 6.9 billion hours sitting in traffic each year [1] -- roughly 10,000 human lifetimes [2]. Autonomous vehicles will be able to alleviate part (but not all) of the problem.

We present a traffic simulation named DeepTraffic where the planning systems for a subset of the vehicles are handled by a neural network as part of a model-free, off-policy reinforcement learning process. The primary goal of DeepTraffic is to make the hands-on study of deep reinforcement learning accessible to thousands of students, educators, and researchers in order to inspire and fuel the exploration and evaluation of deep Q-learning network variants and hyperparameter configurations through large-scale, open competition. This paper investigates the crowd-sourced hyperparameter tuning of the policy network that resulted from the first iteration of the DeepTraffic competition where thousands of participants actively searched through the hyperparameter space.