Super-convergence and Differential Privacy: Training faster with better privacy guarantees

Osvald Frisk; Friedrich   Dormann; Christian Marius Lillelund; Christian Fischer Pedersen

Super-convergence and Differential Privacy: Training faster with better privacy guarantees

Osvald Frisk, Friedrich Dormann, Christian Marius Lillelund, Christian Fischer Pedersen

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

Original language	English
Title of host publication	55th Annual Conference on Information Sciences and Systems
Number of pages	6
Publisher	IEEE
Publication date	2021
Pages	1-6
ISBN (Print)	978-1-6654-4844-4
ISBN (Electronic)	978-1-6654-1268-1
Publication status	Published - 2021
Event	55th Annual Conference on Information Sciences and Systems - Virtual, Baltimore, United States Duration: 24 Mar 2021 → 26 Mar 2021 https://ciss.jhu.edu/

Conference

Conference	55th Annual Conference on Information Sciences and Systems
Location	Virtual
Country/Territory	United States
City	Baltimore
Period	24/03/2021 → 26/03/2021
Internet address	https://ciss.jhu.edu/

Keywords

Machine Learning, Differential Privacy, Rényi Differential Privacy, Deep Learning, Super-convergence

Access to Document

https://arxiv.org/pdf/2103.10498.pdf

Cite this

@inproceedings{5aa48f9fe27f4a6caf61f7f818d099d3,

title = "Super-convergence and Differential Privacy: Training faster with better privacy guarantees",

keywords = "Machine Learning, Differential Privacy, R{\'e}nyi Differential Privacy, Deep Learning, Super-convergence",

author = "Osvald Frisk and Friedrich Dormann and Lillelund, {Christian Marius} and Pedersen, {Christian Fischer}",

year = "2021",

language = "English",

isbn = "978-1-6654-4844-4",

pages = "1--6",

booktitle = "55th Annual Conference on Information Sciences and Systems",

publisher = "IEEE",

note = "55th Annual Conference on Information Sciences and Systems ; Conference date: 24-03-2021 Through 26-03-2021",

url = "https://ciss.jhu.edu/",

}

Frisk, O, Dormann, F, Lillelund, CM & Pedersen, CF 2021, Super-convergence and Differential Privacy: Training faster with better privacy guarantees. in 55th Annual Conference on Information Sciences and Systems. IEEE, pp. 1-6, 55th Annual Conference on Information Sciences and Systems, Baltimore, Maryland, United States, 24/03/2021. <https://arxiv.org/pdf/2103.10498.pdf>

TY - GEN

T1 - Super-convergence and Differential Privacy: Training faster with better privacy guarantees

AU - Frisk, Osvald

AU - Dormann, Friedrich

AU - Lillelund, Christian Marius

AU - Pedersen, Christian Fischer

PY - 2021

Y1 - 2021

KW - Machine Learning, Differential Privacy, Rényi Differential Privacy, Deep Learning, Super-convergence

UR - https://doi.org/10.1109/CISS50987.2021.9400274

M3 - Article in proceedings

SN - 978-1-6654-4844-4

SP - 1

EP - 6

BT - 55th Annual Conference on Information Sciences and Systems

PB - IEEE

T2 - 55th Annual Conference on Information Sciences and Systems

Y2 - 24 March 2021 through 26 March 2021

ER -

Super-convergence and Differential Privacy: Training faster with better privacy guarantees

Conference

Keywords

Access to Document

Other files and links

Cite this