UROP Openings

Have a UROP opening you would like to submit?

Please fill out the form.

Submit your UROP opening

Machine Learning Approach for Enhanced Design of Integrated Photonic Devices




2: Mechanical Engineering

Faculty Supervisor:

Brian Anthony

Faculty email:


Apply by:



Robin Singh: robinme@mit.edu

Project Description

In this project proposal, we will explore the application of Machine Learning in physical sciences. Specifically, we will work towards utilizing the powerful field of machine learning and artificial intelligence in the field of integrated photonics. Integrated photonics is one of the fasting growing fields that enable photonic and electronic devices to communicate with one another within an integrated circuit package. The photonic integrated circuits comprise microstructures such as waveguides, ring resonators, gratings and modulators on a chip to perform optical signal processing. In this project, we will explore AI tools to design new photonic microstructures (often called as photonic meta structures) to engineer different micro-optical beam patterns. We will develop a data-driven approach for inverse modeling of meta-optical structures. Inverse modeling comprises predicting the metasurface design parameters based on a given electromagnetic response. The student would work closely with senior Ph.D. students to brainstorm new methodologies in the project, get regular feedback and work towards preparing a journal manuscript towards the end of the project.


1) Exposure to Python and coding in general. 2) Exposure to basics and fundamentals of Machine learning, AI & Deep Learning algorithms 3) Understanding of fundamental physics, optics and wave theory.