Jul 6 – 8, 2021
Europe/Zurich timezone

Learning Partially Known Stochastic Dynamics with Empirical PAC Bayes

Speaker

Manuel Haußmann (Universität Heidelberg)

Description

Neural Stochastic Differential Equations model a dynamical environment with neural nets assigned to their drift and diffusion terms. The high expressive power of their nonlinearity comes at the expense of instability in the identification of the large set of free parameters. This worok presents a recipe to improve the prediction accuracy of such models in three steps: i) accounting for epistemic uncertainty by assuming probabilistic weights, ii) incorporation of partial knowledge on the state dynamics, and iii) training the resultant hybrid model by an objective derived from a PAC-Bayesian generalization bound. We observe in our experiments that this recipe effectively translates partial and noisy prior knowledge into an improved model fit.

Affiliation Heidelberg University
Academic Rank PhD student

Primary author

Manuel Haußmann (Universität Heidelberg)

Co-authors

Sebastian Gerwinn (Bosch Center for Artificial Intelligence) Andreas Look (Bosch Center for Artificial Intelligence) Barbara Rakitsch (Bosch Center for Artificial Intelligence) Melih Kandemir (Bosch Center for Artificial Intelligence)

Presentation materials