Have a UROP opening you would like to submit?
Please fill out the form.
Deep reinforcement learning and its application in urban mobility
11: Urban Studies and Planning
Peyman Noursalehi, firstname.lastname@example.org
The overarching goal of this project is to understand the effectiveness of deep reinforcement learning techniques in modeling the operation of ridehailing services. This includes real-time pricing of the service and the reward programs for drivers. One of the goals of the project is to leverage multi-agent RL for hierarchical, cooperative policy design, as well as exploring the impact of communication protocols among drivers on the overall efficiency. We are looking for a candidate who is interested in applications at the intersection of machine learning methods and transportation. The ideal candidate should have taken the introduction to machine learning class (6.036), know some RL techniques (e.g., knows about MDP, Gym), object oriented programming, and be proficient with basic data analysis skills by using Python. The task will involve extending a mobility simulation written in Python, writing unit-tests, documenting, and modifying the API as required by the Ray library.
Object oriented programming with Python. Experience with writing unit tests. Basic familiarity with RL and ML