papers AI Learner
The Github is limit! Click to go to the new site.

DeepTraffic: Crowdsourced Hyperparameter Tuning of Deep Reinforcement Learning Systems for Multi-Agent Dense Traffic Navigation

2019-01-03
Lex Fridman, Jack Terwilliger, Benedikt Jenik

Abstract

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.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1801.02805

PDF

http://arxiv.org/pdf/1801.02805


Similar Posts

Comments