Hitting the gym: controlling traffic with Reinforcement Learning
Finally a good real-life use case for Reinforcement Learning (RL): traffic control! In this talk I will show you how we hooked up traffic simulation software to Python and how we built our own custom `gym` environment to run RL experiments with `keras-rl` on a simple 4-way intersection.
The session will start with a conceptual understanding of Reinforcement Learning and how algorithms like (Deep) Q-learning work. I will then explain why this is relevant for traffic control, after which I will zoom in on OpenAI and how to build your own custom `gym` environment. With such an environment you can easily tap into existing `keras-rl` algorithms, which will speed up your RL project significantly.
In our case, connecting the `gym` to the traffic simulation software wasn’t trivial. Therefore, there is also a short note on the use of `multiprocessing` and blocking queues to enable the reinforcement learning agent to gain control of the simulation software.
Data Science Consultant, GoDataDriven
Steven likes to combine thorough data analysis with strategic business thinking. Being a data science consultant since 2014 he has demonstrated the added value of analytics in several industries now, often redefining business processes using data-driven insights. To make his point Steven masters not only R and Python, but also good old PowerPoint.
Before becoming a consultant, Steven studied Econometrics at the University of Amsterdam and worked in the financial industry. He decided to become a data scientist after investigating how sentiment derived from social media text data influences financial markets. His work got published in the Journal of Behavioral Finance. Although the algorithm didn’t make him rich, he gained valuable experience in modelling and programming. Something that he now teaches to clients at GoDataDriven.
Steven is a data and travel enthousiast. In his spare time he combines both passions by building a recommendation engine for travel inspiration. Need a suggestion for your next trip? Check out his hobby project Stairway to Travel.