EECS Colloquium: When should we make our models continuous in time?

306 (HP Auditorium) Soda Hall
  • David Duvenaud, University of Toronto
The world is a continuous-time latent variable model, which makes such models a natural fit for all kinds of messy scientific data that comes at irregular intervals, such as patient records or astronomical observations. It has recently become more practical to use more sophisticated versions of such models, by using neural networks to aid in approximate inference. But practical use of such models...