PAC-Bayesian theory for stochastic LTI systems

Deividas Eringis; John-Josef Leth; Zheng-Hua Tan; Rafal Wisniewski; Alireza Fakhrizadeh Esfahani; Mihaly Petreczky

doi:10.1109/CDC45484.2021.9682808

PAC-Bayesian theory for stochastic LTI systems

Deividas Eringis, John-Josef Leth, Zheng-Hua Tan, Rafal Wisniewski, Alireza Fakhrizadeh Esfahani, Mihaly Petreczky

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

3 Citations (Scopus)

Abstract

In this paper we derive a PAC-Bayesian error bound for autonomous stochastic LTI state-space models. The motivation for deriving such error bounds is that they will allow deriving similar error bounds for more general dynamical systems, including recurrent neural networks. In turn, PAC-Bayesian error bounds are known to be useful for analyzing machine learning algorithms and for deriving new ones.

Original language	English
Title of host publication	2021 60th IEEE Conference on Decision and Control (CDC)
Number of pages	8
Publisher	IEEE
Publication date	2021
Pages	6626-6633
Article number	9682808
ISBN (Print)	978-1-6654-3660-1
ISBN (Electronic)	978-1-6654-3659-5
DOIs	https://doi.org/10.1109/CDC45484.2021.9682808
Publication status	Published - 2021
Event	2021 60th IEEE Conference on Decision and Control (CDC) - Austin, United States Duration: 14 Dec 2021 → 17 Dec 2021

Conference

Conference	2021 60th IEEE Conference on Decision and Control (CDC)
Country/Territory	United States
City	Austin
Period	14/12/2021 → 17/12/2021

Series	I E E E Conference on Decision and Control. Proceedings
ISSN	0743-1546

Access to Document

10.1109/CDC45484.2021.9682808

https://arxiv.org/pdf/2103.12866.pdf

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{41f14ea790bf4ab0a2e153e453036d2b,

title = "PAC-Bayesian theory for stochastic LTI systems",

abstract = "In this paper we derive a PAC-Bayesian error bound for autonomous stochastic LTI state-space models. The motivation for deriving such error bounds is that they will allow deriving similar error bounds for more general dynamical systems, including recurrent neural networks. In turn, PAC-Bayesian error bounds are known to be useful for analyzing machine learning algorithms and for deriving new ones.",

author = "Deividas Eringis and John-Josef Leth and Zheng-Hua Tan and Rafal Wisniewski and {Fakhrizadeh Esfahani}, Alireza and Mihaly Petreczky",

year = "2021",

doi = "10.1109/CDC45484.2021.9682808",

language = "English",

isbn = "978-1-6654-3660-1",

series = "I E E E Conference on Decision and Control. Proceedings",

publisher = "IEEE",

pages = "6626--6633",

booktitle = "2021 60th IEEE Conference on Decision and Control (CDC)",

address = "United States",

note = "2021 60th IEEE Conference on Decision and Control (CDC) ; Conference date: 14-12-2021 Through 17-12-2021",

}

Eringis, D , Leth, J-J , Tan, Z-H , Wisniewski, R, Fakhrizadeh Esfahani, A & Petreczky, M 2021, PAC-Bayesian theory for stochastic LTI systems. in 2021 60th IEEE Conference on Decision and Control (CDC)., 9682808, IEEE, I E E E Conference on Decision and Control. Proceedings, pp. 6626-6633, 2021 60th IEEE Conference on Decision and Control (CDC), Austin, Texas, United States, 14/12/2021. https://doi.org/10.1109/CDC45484.2021.9682808

PAC-Bayesian theory for stochastic LTI systems. / Eringis, Deividas ; Leth, John-Josef ; Tan, Zheng-Hua et al.
2021 60th IEEE Conference on Decision and Control (CDC). IEEE, 2021. p. 6626-6633 9682808 (I E E E Conference on Decision and Control. Proceedings).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - PAC-Bayesian theory for stochastic LTI systems

AU - Eringis, Deividas

AU - Leth, John-Josef

AU - Tan, Zheng-Hua

AU - Wisniewski, Rafal

AU - Fakhrizadeh Esfahani, Alireza

AU - Petreczky, Mihaly

PY - 2021

Y1 - 2021

N2 - In this paper we derive a PAC-Bayesian error bound for autonomous stochastic LTI state-space models. The motivation for deriving such error bounds is that they will allow deriving similar error bounds for more general dynamical systems, including recurrent neural networks. In turn, PAC-Bayesian error bounds are known to be useful for analyzing machine learning algorithms and for deriving new ones.

AB - In this paper we derive a PAC-Bayesian error bound for autonomous stochastic LTI state-space models. The motivation for deriving such error bounds is that they will allow deriving similar error bounds for more general dynamical systems, including recurrent neural networks. In turn, PAC-Bayesian error bounds are known to be useful for analyzing machine learning algorithms and for deriving new ones.

U2 - 10.1109/CDC45484.2021.9682808

DO - 10.1109/CDC45484.2021.9682808

M3 - Article in proceeding

SN - 978-1-6654-3660-1

T3 - I E E E Conference on Decision and Control. Proceedings

SP - 6626

EP - 6633

BT - 2021 60th IEEE Conference on Decision and Control (CDC)

PB - IEEE

T2 - 2021 60th IEEE Conference on Decision and Control (CDC)

Y2 - 14 December 2021 through 17 December 2021

ER -

PAC-Bayesian theory for stochastic LTI systems

Abstract

Conference

Access to Document

AUB Link

Fingerprint

Cite this