Surviving Solver Sensitivity: An ASP Practitioner’s Guide

Authors Bryan Silverthorn, Yuliya Lierler, Marius Schneider



PDF
Thumbnail PDF

File

LIPIcs.ICLP.2012.164.pdf
  • Filesize: 0.49 MB
  • 12 pages

Document Identifiers

Author Details

Bryan Silverthorn
Yuliya Lierler
Marius Schneider

Cite AsGet BibTex

Bryan Silverthorn, Yuliya Lierler, and Marius Schneider. Surviving Solver Sensitivity: An ASP Practitioner’s Guide. In Technical Communications of the 28th International Conference on Logic Programming (ICLP'12). Leibniz International Proceedings in Informatics (LIPIcs), Volume 17, pp. 164-175, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2012)
https://doi.org/10.4230/LIPIcs.ICLP.2012.164

Abstract

Answer set programming (ASP) is a declarative programming formalism that allows a practitioner to specify a problem without describing an algorithm for solving it. In ASP, the tools for processing problem specifications are called answer set solvers. Because specified problems are often NP complete, these systems often require significant computational effort to succeed. Furthermore, they offer different heuristics, expose numerous parameters, and their running time is sensitive to the configuration used. Portfolio solvers and automatic algorithm configuration systems are recent attempts to automate the problem of manual parameter tuning, and to mitigate the burden of identifying the right solver configuration. The approaches taken in portfolio solvers and automatic algorithm configuration systems are orthogonal. This paper evaluates these approaches, separately and jointly, in the context of real-world ASP application development. It outlines strategies for their use in such settings, identifies their respective strengths and weaknesses, and advocates for a methodology that would make them an integral part of developing ASP applications.
Keywords
  • algorithm configuration
  • algorithm selection
  • portfolio solving
  • answer set programming
  • algorithm portfolios

Metrics

  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    0
    PDF Downloads
Questions / Remarks / Feedback
X

Feedback for Dagstuhl Publishing


Thanks for your feedback!

Feedback submitted

Could not send message

Please try again later or send an E-mail