Challenges of Program Synthesis with Grammatical Evolution

Publikationstyp:
Konferenzbeitrag
Metadaten:
Autoren
DOI
10.1007/978-3-030-44094-7_14
eISSN
1611-3349
ISBN-13
9783030440930
ISSN
0302-9743
Online publication date
2020
Paginierung
211 - 227
Datum der Veröffentlichung
2020
Status
Published
Herausgeber
Springer International Publishing
Herausgeber URL
http://dx.doi.org/10.1007/978-3-030-44094-7_14
Datum der Datenerfassung
2021
Titel
Challenges of Program Synthesis with Grammatical Evolution

Datenquelle: Crossref

Andere Metadatenquellen:
Autoren
Editoren
  • Ting Hu
  • Nuno Lourenço
  • Eric Medvet
  • Federico Divina
ISBN-13
978-3-030-44093-0
Zeitschrift
EuroGP
Paginierung
211 - 227
Datum der Veröffentlichung
2020
Herausgeber
Springer
Herausgeber URL
https://doi.org/10.1007/978-3-030-44094-7
Titel
Challenges of Program Synthesis with Grammatical Evolution.
Ausgabe der Zeitschrift
12101

Datenquelle: DBLP

Abstract
Program synthesis is an emerging research topic in the field of EC with the potential to improve real-world software development. Grammar-guided approaches like GE are suitable for program synthesis as they can express common programming languages with their required properties. This work uses common software metrics (lines of code, McCabe metric, size and depth of the abstract syntax tree) for an analysis of GE’s search behavior and the resulting problem structure. We find that GE is not able to solve program synthesis problems, where correct solutions have higher values of the McCabe metric (which means they require conditions or loops). Since small mutations of high-quality solutions strongly decrease a solution’s fitness and make a high percentage of the solutions non-executable, the resulting problem constitutes a needle-in-a-haystack problem. To us, one of the major challenges of future GP research is to come up with better and more adequate fitness functions and problem specifications to turn the current needle-in-a-haystack problems into problems that can be solved by guided search.
Date of acceptance
2020
Autoren
DOI
10.1007/978-3-030-44094-7
Conference finish date
2020
ISBN-13
978-3-030-44093-0
ISSN
0302-9743
Zeitschrift
Proceedings of 23rd European Conference on Genetic Programming
Conference place
Sevilla, Spain
Name of conference
EURO GP (23rd European Conference on Genetic Programming)
Paginierung
211 - 227
Ort der Veröffentlichung
Heidelberg
Datum der Veröffentlichung
2020
Status
Published
Herausgeber
Springer LNCS
Datum der Datenerfassung
2020
Conference start date
2020
Titel
Challenges of Program Synthesis with Grammatical Evolution

Datenquelle: Manual

Beziehungen:
Eigentum von

Werkzeuge