Skip to Main Content

Stephanie Milani

“I primarily chose UMBC due to financial reasons, but, once I became more involved on campus, I grew to love the opportunities, resources, and communities that I had previously overlooked.”

Major: Computer Science
Major: Psychology
Certificate: Honors College

Hometown: Montreal, QC, Canada

Campus Activities:
President of the Computer Science Education Club; Member of the Russian Club; Curriculum Development Coordinator and Co-founder of Creative Coders; Shriver Center Student Coordinator; National Academy of Engineering Grand Challenge Scholar (G2); 2017-2018 France-Merrick Scholar; 2018-2019 Newman Civic Fellow; 2018-2019 URA Scholar; CWIT Affiliate; Cyber Scholars Affiliate; Association for Computing Machinery (ACM). Current research in Dr. Marie desJardins’ laboratory in the Computer Science and Electrical Engineering department. Previous research internships at Carnegie Mellon University (deep learning and computer vision) and the University of Maryland, School of Medicine (neurobiology). Previous internship with Computer Science Matters in Maryland.

Why did you chose UMBC?
I primarily chose UMBC due to financial reasons, but, once I became more involved on campus, I grew to love the opportunities, resources, and communities that I had previously overlooked.

What has been your favorite class at UMBC?
It’s difficult to pick a favorite class because there are so many great professors and courses at UMBC! If I had to choose just one class, it would probably be CMSC 341: Data Structures with Professor Park. He is a great professor who is invested in the success of his students. The projects for the course were challenging, but I learned a lot from them. Professor Park also makes some of the most fun exams I have ever taken at UMBC: many of the questions were like fun logic puzzles!

What are you researching?
I work in Dr. Marie desJardins’ laboratory in the Computer Science and Electrical Engineering department. My research focuses on model-based learning for abstract Markov decision processes (AMDPs), an approach by which agents can learn to decompose complex planning problems into a hierarchy of distinct, related subtasks that use a widely-known problem representation for reinforcement learning. The spring of 2019 will mark my first time participating in URCAD, where I will present my URA-funded work of integrating ethical reasoning into the AMDP framework.

Where did you complete your internship/applied learning experience?
Last summer, I was a Robotics Institute Summer Scholar at Carnegie Mellon University in Pittsburgh, PA, where I worked in the Navlab under the guidance of Dr. Christoph Mertz. There were 35 students selected from around the world. During my stay, I assembled all of the hardware and software components of a deep learning system and researched using an adversarial neural network to improve the detection of occluded traffic signs in a road infrastructure inventory and assessment system.

What do you hope to achieve after you complete your degree at UMBC?
After I graduate, I plan to pursue a PhD in Artificial Intelligence or Machine Learning.

More Student Profiles