With a PhD in stats, University of Toronto grad Cédric Beaulac plans to teach, do research and explore video game design
University of Toronto graduate Cédric Beaulac managed to land his dream job in academia, but he still plans to find a way to pursue “geek stuff” aspirations, including designing video games.
After earning his PhD in statistical sciences in the Faculty of Arts & Science, Beaulac will spend a year as a CANSSI (Canadian Statistical Sciences Institute) distinguished post-doctoral researcher at Simon Fraser University and the University of Victoria.
An award-winning teaching assistant at U of T, he will then join the faculty of Université du Québec à Montréal (UQAM) in his hometown as an associate professor in January of 2023.
Between the end of his post-doc and beginning of his career as a university professor, Beaulac also plans to complete a six-month internship at a video game company.
For Beaulac, it’s the perfect way to dip his toes into the tech industry while exploring the link between developing algorithms for statistical sciences and generating creative content for video games and movies.
Writer Peter Boisseau recently spoke to Beaulac about his statisical sciences research, how it can be used to create trees and other background objects in films like Star Wars – and what the future holds.
Your recent research has focused on survival analysis, time series and content generation. Could you offer everyday examples of those things?
The most common application for survival analysis is in medicine, where we try to assess the efficacy of a treatment by studying a variable in the data set for a group of patients. Time series is more closely related to economics. For example, I’ve done time series projects looking at electricity demand in Ontario, assessing how to ensure the peaks in demand do not overtax the supply or system. Content generation is something new for me. I’m trying to explore how algorithms could be used to create sets of art assets such as graphics or even music in a very inexpensive manner.
Does content generation dovetail with your passion for video games in any way?
Very much so. I like sciences and most geek stuff, and, if I could duplicate myself, the other me would definitely study video game design. Content generation is something I am going to explore in a post-doc, and I’d like to eventually make it the centre of my career. The idea is to understand the pattern and randomness of natural events and replicate them as best as possible. One of the first to employ it was a Pixar movie where they did a wooden floor using content generation algorithms, without having to pay a graphic engineer to manually design those wooden planks one by one. It’s not meant to replace artists, but to empower them. There’s a company that’s been making trees in video games for years, and they’ve recently made a big breakthrough in movies. In the latest Star Wars, all the trees are randomly generated by their particular program.
What impact did U of T have on your career path?
U of T is a very competitive environment for research. I think it really helped me figure out the kind of environment I want to do my research in. I think it also made me realize I’m really going to enjoy the teaching aspect of being a professor, and I’m even willing to sacrifice research time to make room for my teaching. For me, it’s an important part of being a professor. I worked under the supervision of Professor Jeffrey Rosenthal, and he was very influential on me. Whenever my confidence would slip or I would get discouraged, he’d remind me this is a field that takes time and practice. When I start teaching at UQAM, I’d like to be a bit like him.
What advice would you offer to students?
To undergraduate students I’d say, focus on learning over grades. Grades and diplomas aren’t true goals, but understanding and learning should be. To PhD students I would say to focus on what you enjoy above all and don’t think too much about careers and outside pressure. You’ll be working on your own, reading papers for hours. Enjoying your work is definitely the most important thing.