UROP Openings

Have a UROP opening you would like to submit?

Please fill out the form.

Submit your UROP opening

Reinforcement Learning Model Development and Comparison


Term:

Fall

Department:

4: Architecture

Faculty Supervisor:

Takehiko Nagakura

Faculty email:

takehiko@mit.edu

Apply by:

September 15 (13 to prepare funding application)

Contact:

Paloma Gonzalez: palomagr@mit.edu

Project Description

We are using human trajectory data to feed a Reinforcement Learning Model in Unity3D. We have two training approaches to compare, Exploratory RL and Imitation Learning. For each a training scene is set up with rewards and a curated label system. Such scenes have been set up, and the main task is to test them in order to do a quantitative/qualitative comparison. 3 positions available.

Pre-requisites

C# and Python programming skills in medium level, object oriented programming, Reinforcement Learning experience / knowledge Unity3D experience