Genetic Configuration Sampling: Learning a Sampling Strategy for Fault Detection of Configurable Systems
Created by W.Langdon from
gp-bibliography.bib Revision:1.7970
- @InProceedings{Xuan:2018:GI5,
-
author = "Jifeng Xuan and Yongfeng Gu and Zhilei Ren and
Xiangyang Jia and Qingna Fan",
-
title = "Genetic Configuration Sampling: Learning a Sampling
Strategy for Fault Detection of Configurable Systems",
-
booktitle = "5th edition of GI @ GECCO 2018",
-
year = "2018",
-
editor = "Brad Alexander and Saemundur O. Haraldsson and
Markus Wagner and John R. Woodward and Shin Yoo",
-
pages = "1624--1631",
-
address = "Kyoto, Japan",
-
month = "15-19 " # jul,
-
organisation = "ACM SIGEvo",
-
publisher = "ACM",
-
keywords = "genetic algorithms, genetic programming, genetic
improvement, SBSE, Configuration sampling, fault
detection, highly-configurable systems, software
configurations",
-
URL = "http://www.cs.stir.ac.uk/events/gecco-gi-2018/papers/genetic_configuration_sampling.pdf",
-
DOI = "doi:10.1145/3205651.3208267",
-
size = "8 pages",
-
abstract = "A highly-configurable system provides many
configuration options to diversify application
scenarios. The combination of these configuration
options results in a large search space of
configurations. This makes the detection of
configuration-related faults extremely hard. Since it
is infeasible to exhaust every configuration, several
methods are proposed to sample a subset of all
configurations to detect hidden faults. Configuration
sampling can be viewed as a process of repeating a
pre-defined sampling action to the whole search space,
such as the one-enabled or pair-wise strategy.
we propose genetic configuration sampling, a new method
of learning a sampling strategy for
configuration-related faults. Genetic configuration
sampling encodes a sequence of sampling actions as a
chromosome in the genetic algorithm. Given a set of
known configuration-related faults, genetic
configuration sampling evolves the sequence of sampling
actions and applies the learnt sequence to new
configuration data. A pilot study on three
highly-configurable systems shows that genetic
configuration sampling performs well among nine
sampling strategies in comparison.",
-
notes = "Apache, BusyBox, Linux
'compiler Gcc 7.3 contains 2472 configuration
options'
http://www.cs.stir.ac.uk/events/gecco-gi-2018/cfp.html",
- }
Genetic Programming entries for
Jifeng Xuan
Yongfeng Gu
Zhilei Ren
Xiangyang Jia
Qingna Fan
Citations