DeepTraffic: Crowdsourced Hyperparameter Tuning of Deep Reinforcement Learning Systems for Multi-Agent Dense Traffic Navigation
This work addresses the challenge of hyperparameter optimization for deep reinforcement learning in traffic simulation, making it accessible for education and research, but it is incremental as it applies existing methods to a new platform.
The authors tackled the problem of tuning deep reinforcement learning hyperparameters for multi-agent traffic navigation by crowdsourcing through an open competition, resulting in thousands of participants exploring the hyperparameter space.
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.