Bayesian learning via neural Schrödinger–Föllmer flows
-
Published:2022-11-23
Issue:1
Volume:33
Page:
-
ISSN:0960-3174
-
Container-title:Statistics and Computing
-
language:en
-
Short-container-title:Stat Comput
Author:
Vargas FranciscoORCID, Ovsianas Andrius, Fernandes David, Girolami Mark, Lawrence Neil D., Nüsken Nikolas
Abstract
AbstractIn this work we explore a new framework for approximate Bayesian inference in large datasets based on stochastic control. We advocate stochastic control as a finite time and low variance alternative to popular steady-state methods such as stochastic gradient Langevin dynamics. Furthermore, we discuss and adapt the existing theoretical guarantees of this framework and establish connections to already existing VI routines in SDE-based models.
Funder
Engineering and Physical Sciences Research Council Deutsche Forschungsgemeinschaft Huawei Technologies
Publisher
Springer Science and Business Media LLC
Subject
Computational Theory and Mathematics,Statistics, Probability and Uncertainty,Statistics and Probability,Theoretical Computer Science
Reference78 articles.
1. Amari, S.-i., Cichocki, A., Yang, H. H., et al.: A new learning algorithm for blind signal separation. In: Advances in Neural Information Processing Systems, pp. 757–763. Morgan Kaufmann Publishers (1996) 2. Bartholomew-Biggs, M., Brown, S., Christianson, B., Dixon, L.: Automatic differentiation of algorithms. J. Comput. Appl. Math. 124(1–2), 171–190 (2000) 3. Belkin, M., Hsu, D., Ma, S., Mandal, S.: Reconciling modern machine-learning practice and the classical bias-variance trade-off. Proc. Natl. Acad. Sci. 116(32), 15849–15854 (2019) 4. Bernton, E., Heng, J., Doucet, A., Jacob, P.E.: Schrödinger bridge samplers (2019). arXiv preprint 5. Bioucas-Dias, J.M., Plaza, A., Dobigeon, N., Parente, M., Du, Q., Gader, P., Chanussot, J.: Hyperspectral unmixing overview: geometrical, statistical, and sparse regression-based approaches. IEEE J. Select. Topics Appl. Earth Obs. Remote Sens. 5(2), 354–379 (2012)
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献
|
|