log in  |  register  |  feedback?  |  help  |  web accessibility
Differentiable Physics: Computer Graphics as an Inductive Bias
Derek Nowrouzezahrai - McGill University
IRB 4105
Tuesday, October 12, 2021, 3:30-5:00 pm Calendar
  • You are subscribed to this talk through .
  • You are watching this talk through .
  • You are subscribed to this talk. (unsubscribe, watch)
  • You are watching this talk. (unwatch, subscribe)
  • You are not subscribed to this talk. (watch, subscribe)

Also available on zoom- https://umd.zoom.us/j/4603492472

After briefly touring the increasing interplay between computer graphics and machine learning, I will highlight one recent path of growing intersection between these two communities: the development and application of differentiable simulators.
Here, I make the specific claim that expertise and advances development in the computational physics, statistical and numerical methods, and — ultimately — computer graphics communities can lead to powerful inductive biases in physics-oriented learning tasks.  I will survey the many reasons — motivated in part by a recent case study — for my excitement in maintaining and a research agenda that combines computer graphics-oriented methodologies to physics-based machine learning.


Derek Nowrouzezahrai is an Associate Professor at McGill University and a Core Faculty Member of the Quebec Institute for Artificial Intelligence (Mila). He Directs the McGill Graphics Lab and McGill University's Centre for Intelligent Machines, a grouping of over a dozen labs in the Faculty of Engineering and the Faculty of Science. He held a Post-Doctoral position at Disney Research Zurich after completing his graduate work at the University of Toronto.

Derek’s group works on devising new mathematical models of visual phenomena and dynamics, developing efficient and differentiable numerical approaches to solve forward and inverse physics-based problems. These include problems in light transport, fluid dynamics and control, robotics, augmented reality, digital manufacturing, computational optics and imaging, and image and geometry processing.  Derek's research has been adopted in feature films, video games, amusement parks and consumer products. For more information, please refer to his website at https://www.cim.mcgill.ca/~derek/
This talk is organized by Richa Mathur