We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Differential Bayesian Neural Nets

Abstract: Neural Ordinary Differential Equations (N-ODEs) are a powerful building block for learning systems, which extend residual networks to a continuous-time dynamical system. We propose a Bayesian version of N-ODEs that enables well-calibrated quantification of prediction uncertainty, while maintaining the expressive power of their deterministic counterpart. We assign Bayesian Neural Nets (BNNs) to both the drift and the diffusion terms of a Stochastic Differential Equation (SDE) that models the flow of the activation map in time. We infer the posterior on the BNN weights using a straightforward adaptation of Stochastic Gradient Langevin Dynamics (SGLD). We illustrate significantly improved stability on two synthetic time series prediction tasks and report better model fit on UCI regression benchmarks with our method when compared to its non-Bayesian counterpart.
Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
Journal reference: 4th workshop on Bayesian Deep Learning (NeurIPS 2019), Vancouver, Canada
Cite as: arXiv:1912.00796 [cs.LG]
  (or arXiv:1912.00796v2 [cs.LG] for this version)

Submission history

From: Andreas Look [view email]
[v1] Mon, 2 Dec 2019 14:03:55 GMT (885kb,D)
[v2] Tue, 18 Feb 2020 10:14:31 GMT (885kb,D)

Link back to: arXiv, form interface, contact.