CCM Colloquium / Group Meeting

CCM Seminar: David Duvenaud (U Toronto)

America/New_York
Description

Title: Latent Stochastic Differential Equations: An Unexplored Model Class.

Abstract: We show how to do gradient-based stochastic variational inference in stochastic differential equations (SDEs), in a way that allows the use of adaptive SDE solvers.  This allows us to scalably fit a new family of richly-parameterized distributions over irregularly-sampled time series.  We apply latent SDEs to motion capture data, and to demonstrate infinitely-deep Bayesian neural networks.  We also discuss the pros and cons of this barely-explored model class, comparing it to Gaussian processes and neural processes.

Some technical details are in this paper: https://arxiv.org/abs/2001.01328
And code is available at: https://github.com/google-research/torchsde
duvenaud@cs.toronto.edu