SuperSCS: fast and accurate large-scale conic optimization

Pantelis Sopasakis; Krina Menounou; Panagiotis Patrinos

doi:10.23919/ECC.2019.8796286

SuperSCS: fast and accurate large-scale conic optimization

Pantelis Sopasakis, Krina Menounou, Panagiotis Patrinos

School of Electronics, Electrical Engineering and Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

13 Citations (Scopus)

530 Downloads (Pure)

Abstract

We present SuperSCS: a fast and accurate method for solving large-scale convex conic problems. SuperSCS combines the SuperMann algorithmic framework with the Douglas-Rachford splitting which is applied on the homogeneous self-dual embedding of conic optimization problems: a model for conic optimization problems which simultaneously encodes the optimality conditions and infeasibility/unboundedness certificates for the original problem. SuperMann allows the use of fast quasi-Newtonian directions such as a modified restarted Broyden-type direction and Anderson's acceleration.

Original language	English
Title of host publication	European Control Conference 25/06/2018 → 28/06/2019 Naples, Italy
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1500-1505
Number of pages	6
ISBN (Electronic)	978-3-907144-00-8
ISBN (Print)	978-1-7281-1314-2
DOIs	https://doi.org/10.23919/ECC.2019.8796286
Publication status	Published - 15 Aug 2019
Event	European Control Conference - Hotel Royal Continental, Naples, Naples, Italy Duration: 25 Jun 2018 → 28 Jun 2019 https://ecc19.eu/

Conference

Conference	European Control Conference
Abbreviated title	ECC
Country/Territory	Italy
City	Naples
Period	25/06/2018 → 28/06/2019
Internet address	https://ecc19.eu/

Keywords

Conic optimization
Convex optimization
Numerical Optimization
Numerical analysis
Homogeneous self-dual embedding
Large-scale optimization

Access to Document

10.23919/ECC.2019.8796286Licence: Unspecified

SuperSCS: fast and accurate large-scale conic optimization
Copyright 2019 IEEE. This work is made available online in accordance with the publisher’s policies. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 410 KBLicence: Unspecified

Cite this

@inproceedings{7a0912fcd7d84fa6ad1ce5c2ed6c472f,

title = "SuperSCS: fast and accurate large-scale conic optimization",

abstract = "We present SuperSCS: a fast and accurate method for solving large-scale convex conic problems. SuperSCS combines the SuperMann algorithmic framework with the Douglas-Rachford splitting which is applied on the homogeneous self-dual embedding of conic optimization problems: a model for conic optimization problems which simultaneously encodes the optimality conditions and infeasibility/unboundedness certificates for the original problem. SuperMann allows the use of fast quasi-Newtonian directions such as a modified restarted Broyden-type direction and Anderson's acceleration.",

keywords = "Conic optimization, Convex optimization, Numerical Optimization, Numerical analysis, Homogeneous self-dual embedding, Large-scale optimization",

author = "Pantelis Sopasakis and Krina Menounou and Panagiotis Patrinos",

year = "2019",

month = aug,

day = "15",

doi = "10.23919/ECC.2019.8796286",

language = "English",

isbn = "978-1-7281-1314-2",

pages = "1500--1505",

booktitle = "European Control Conference 25/06/2018 → 28/06/2019 Naples, Italy",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "European Control Conference, ECC ; Conference date: 25-06-2018 Through 28-06-2019",

url = "https://ecc19.eu/",

}

TY - GEN

T1 - SuperSCS: fast and accurate large-scale conic optimization

AU - Sopasakis, Pantelis

AU - Menounou, Krina

AU - Patrinos, Panagiotis

PY - 2019/8/15

Y1 - 2019/8/15

N2 - We present SuperSCS: a fast and accurate method for solving large-scale convex conic problems. SuperSCS combines the SuperMann algorithmic framework with the Douglas-Rachford splitting which is applied on the homogeneous self-dual embedding of conic optimization problems: a model for conic optimization problems which simultaneously encodes the optimality conditions and infeasibility/unboundedness certificates for the original problem. SuperMann allows the use of fast quasi-Newtonian directions such as a modified restarted Broyden-type direction and Anderson's acceleration.

AB - We present SuperSCS: a fast and accurate method for solving large-scale convex conic problems. SuperSCS combines the SuperMann algorithmic framework with the Douglas-Rachford splitting which is applied on the homogeneous self-dual embedding of conic optimization problems: a model for conic optimization problems which simultaneously encodes the optimality conditions and infeasibility/unboundedness certificates for the original problem. SuperMann allows the use of fast quasi-Newtonian directions such as a modified restarted Broyden-type direction and Anderson's acceleration.

KW - Conic optimization

KW - Convex optimization

KW - Numerical Optimization

KW - Numerical analysis

KW - Homogeneous self-dual embedding

KW - Large-scale optimization

U2 - 10.23919/ECC.2019.8796286

DO - 10.23919/ECC.2019.8796286

M3 - Conference contribution

SN - 978-1-7281-1314-2

SP - 1500

EP - 1505

BT - European Control Conference 25/06/2018 → 28/06/2019 Naples, Italy

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - European Control Conference

Y2 - 25 June 2018 through 28 June 2019

ER -

SuperSCS: fast and accurate large-scale conic optimization

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this