Discriminating instance generation from abstract specifications: a case study with CP and MIP

Özgür Akgün; Nguyen Dang; Ian Miguel; András Z. Salamon; Patrick Spracklen; Christopher Stone

doi:10.1007/978-3-030-58942-4_3

Discriminating instance generation from abstract specifications: a case study with CP and MIP

Özgür Akgün, Nguyen Dang^*, Ian Miguel, András Z. Salamon, Patrick Spracklen, Christopher Stone

^*Corresponding author for this work

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

Abstract

We extend automatic instance generation methods to allow cross-paradigm comparisons. We demonstrate that it is possible to completely automate the search for benchmark instances that help to discriminate between solvers. Our system starts from a high level human-provided problem specification, which is translated into a specification for valid instances. We use the automated algorithm configuration tool irace to search for instances, which are translated into inputs for both MIP and CP solvers by means of the Conjure, Savile Row, and MiniZinc tools. These instances are then solved by CPLEX and Chuffed, respectively. We constrain our search for instances by requiring them to exhibit a significant advantage for MIP over CP, or vice versa. Experimental results on four optimisation problem classes demonstrate the effectiveness of our method in identifying instances that highlight differences in performance of the two solvers.

Original language	English
Title of host publication	Integration of Constraint Programming, Artificial Intelligence, and Operations Research
Subtitle of host publication	17th International Conference, CPAIOR 2020, Vienna, Austria, September 21–24, 2020, Proceedings
Editors	Emmanuel Hebrard, Nysret Musliu
Place of Publication	Cham
Publisher	Springer
Pages	41-51
Number of pages	11
ISBN (Electronic)	9783030589424
ISBN (Print)	9783030589417
DOIs	https://doi.org/10.1007/978-3-030-58942-4_3
Publication status	Published - 2020
Event	17th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, CPAIOR 2020 - Vienna, Online, Austria Duration: 21 Sept 2020 → 24 Sept 2020 https://cpaior2020.dbai.tuwien.ac.at/

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12296 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	17th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, CPAIOR 2020
Country/Territory	Austria
City	Vienna, Online
Period	21/09/20 → 24/09/20
Internet address	https://cpaior2020.dbai.tuwien.ac.at/

Keywords

Constraint Programming
Instance generation
MIP

Access to Document

10.1007/978-3-030-58942-4_3

Akgün-2020-Discriminating-instance-recognition-CPAIOR-AAM
Copyright © 2020 Springer Nature Switzerland AG. This work has been made available online in accordance with publisher policies or with permission. Permission for further reuse of this content should be sought from the publisher or the rights holder. This is the author created accepted manuscript following peer review and may differ slightly from the final published version. The final published version of this work is available at https://doi.org/10.1007/978-3-030-58942-4_3.
Accepted author manuscript, 1.27 MB

Discriminating instance generation from abstract specifications: a case study with CP and MIP (dataset)
Akgun, O. (Creator), Dang, N. T. T. (Creator) & Salamon, A. (Creator), GitHub, 2020
https://github.com/stacs-cp/CPAIOR2020-InstanceGen
Dataset

Cite this

Akgün, Ö., Dang, N., Miguel, I., Salamon, A. Z., Spracklen, P., & Stone, C. (2020). Discriminating instance generation from abstract specifications: a case study with CP and MIP. In E. Hebrard, & N. Musliu (Eds.), Integration of Constraint Programming, Artificial Intelligence, and Operations Research: 17th International Conference, CPAIOR 2020, Vienna, Austria, September 21–24, 2020, Proceedings (pp. 41-51). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12296 LNCS). Springer. https://doi.org/10.1007/978-3-030-58942-4_3

Akgün, Özgür ; Dang, Nguyen ; Miguel, Ian et al. / Discriminating instance generation from abstract specifications : a case study with CP and MIP. Integration of Constraint Programming, Artificial Intelligence, and Operations Research: 17th International Conference, CPAIOR 2020, Vienna, Austria, September 21–24, 2020, Proceedings. editor / Emmanuel Hebrard ; Nysret Musliu. Cham : Springer, 2020. pp. 41-51 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{04b6f9716a204104b26a8a664d101d92,

title = "Discriminating instance generation from abstract specifications: a case study with CP and MIP",

abstract = "We extend automatic instance generation methods to allow cross-paradigm comparisons. We demonstrate that it is possible to completely automate the search for benchmark instances that help to discriminate between solvers. Our system starts from a high level human-provided problem specification, which is translated into a specification for valid instances. We use the automated algorithm configuration tool irace to search for instances, which are translated into inputs for both MIP and CP solvers by means of the Conjure, Savile Row, and MiniZinc tools. These instances are then solved by CPLEX and Chuffed, respectively. We constrain our search for instances by requiring them to exhibit a significant advantage for MIP over CP, or vice versa. Experimental results on four optimisation problem classes demonstrate the effectiveness of our method in identifying instances that highlight differences in performance of the two solvers.",

keywords = "Constraint Programming, Instance generation, MIP",

author = "{\"O}zg{\"u}r Akg{\"u}n and Nguyen Dang and Ian Miguel and Salamon, {Andr{\'a}s Z.} and Patrick Spracklen and Christopher Stone",

note = "This work is supported by EPSRC grant EP/P015638/1 and used the Cirrus UK National Tier-2 HPC Service at EPCC funded by the University of Edinburgh and EPSRC (EP/P020267/1).; 17th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, CPAIOR 2020 ; Conference date: 21-09-2020 Through 24-09-2020",

year = "2020",

doi = "10.1007/978-3-030-58942-4_3",

language = "English",

isbn = "9783030589417",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "41--51",

editor = "Emmanuel Hebrard and Nysret Musliu",

booktitle = "Integration of Constraint Programming, Artificial Intelligence, and Operations Research",

address = "Netherlands",

url = "https://cpaior2020.dbai.tuwien.ac.at/",

}

Akgün, Ö , Dang, N , Miguel, I , Salamon, AZ, Spracklen, P & Stone, C 2020, Discriminating instance generation from abstract specifications: a case study with CP and MIP. in E Hebrard & N Musliu (eds), Integration of Constraint Programming, Artificial Intelligence, and Operations Research: 17th International Conference, CPAIOR 2020, Vienna, Austria, September 21–24, 2020, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12296 LNCS, Springer, Cham, pp. 41-51, 17th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, CPAIOR 2020, Vienna, Online, Austria, 21/09/20. https://doi.org/10.1007/978-3-030-58942-4_3

Discriminating instance generation from abstract specifications: a case study with CP and MIP. / Akgün, Özgür ; Dang, Nguyen ; Miguel, Ian et al.
Integration of Constraint Programming, Artificial Intelligence, and Operations Research: 17th International Conference, CPAIOR 2020, Vienna, Austria, September 21–24, 2020, Proceedings. ed. / Emmanuel Hebrard; Nysret Musliu. Cham: Springer, 2020. p. 41-51 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12296 LNCS).

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

TY - GEN

T1 - Discriminating instance generation from abstract specifications

T2 - 17th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, CPAIOR 2020

AU - Akgün, Özgür

AU - Dang, Nguyen

AU - Miguel, Ian

AU - Salamon, András Z.

AU - Spracklen, Patrick

AU - Stone, Christopher

N1 - This work is supported by EPSRC grant EP/P015638/1 and used the Cirrus UK National Tier-2 HPC Service at EPCC funded by the University of Edinburgh and EPSRC (EP/P020267/1).

PY - 2020

Y1 - 2020

N2 - We extend automatic instance generation methods to allow cross-paradigm comparisons. We demonstrate that it is possible to completely automate the search for benchmark instances that help to discriminate between solvers. Our system starts from a high level human-provided problem specification, which is translated into a specification for valid instances. We use the automated algorithm configuration tool irace to search for instances, which are translated into inputs for both MIP and CP solvers by means of the Conjure, Savile Row, and MiniZinc tools. These instances are then solved by CPLEX and Chuffed, respectively. We constrain our search for instances by requiring them to exhibit a significant advantage for MIP over CP, or vice versa. Experimental results on four optimisation problem classes demonstrate the effectiveness of our method in identifying instances that highlight differences in performance of the two solvers.

AB - We extend automatic instance generation methods to allow cross-paradigm comparisons. We demonstrate that it is possible to completely automate the search for benchmark instances that help to discriminate between solvers. Our system starts from a high level human-provided problem specification, which is translated into a specification for valid instances. We use the automated algorithm configuration tool irace to search for instances, which are translated into inputs for both MIP and CP solvers by means of the Conjure, Savile Row, and MiniZinc tools. These instances are then solved by CPLEX and Chuffed, respectively. We constrain our search for instances by requiring them to exhibit a significant advantage for MIP over CP, or vice versa. Experimental results on four optimisation problem classes demonstrate the effectiveness of our method in identifying instances that highlight differences in performance of the two solvers.

KW - Constraint Programming

KW - Instance generation

KW - MIP

U2 - 10.1007/978-3-030-58942-4_3

DO - 10.1007/978-3-030-58942-4_3

M3 - Conference contribution

AN - SCOPUS:85092159328

SN - 9783030589417

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 41

EP - 51

BT - Integration of Constraint Programming, Artificial Intelligence, and Operations Research

A2 - Hebrard, Emmanuel

A2 - Musliu, Nysret

PB - Springer

CY - Cham

Y2 - 21 September 2020 through 24 September 2020

ER -

Akgün Ö , Dang N , Miguel I , Salamon AZ, Spracklen P, Stone C. Discriminating instance generation from abstract specifications: a case study with CP and MIP. In Hebrard E, Musliu N, editors, Integration of Constraint Programming, Artificial Intelligence, and Operations Research: 17th International Conference, CPAIOR 2020, Vienna, Austria, September 21–24, 2020, Proceedings. Cham: Springer. 2020. p. 41-51. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Epub 2020 Sept 19. doi: 10.1007/978-3-030-58942-4_3

Discriminating instance generation from abstract specifications: a case study with CP and MIP

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Datasets

Discriminating instance generation from abstract specifications: a case study with CP and MIP (dataset)

Cite this