Screening the parameters affecting heuristic performance

Enda Ridge; Daniel Kudenko

doi:10.1145/1276958.1276994

Screening the parameters affecting heuristic performance

Enda Ridge, Daniel Kudenko

Computer Science

Research output: Contribution to conference › Paper › peer-review

Abstract

This research screens the tuning parameters of a combinatorial optimization heuristic. Specifically, it presents a Design of Experiments (DOE) approach that uses a Fractional Factorial Design to screen the tuning parameters of Ant Colony System (ACS) for the Travelling Sales person problem. Screening is a preliminary step towards building a full Response Surface Model (RSM) [2]. It identifies parametersthat have little influence on performance and can be omittedfrom the RSM design. This reduces the complexity andexpense of the RSM design. 10 algorithm parameters and 2 problem characteristics are considered. Open questionson the effect of 3 parameters on performance are answered.A further parameter, sometimes assumed important, was shown to have no effect on performance. A new problem characteristic that effects performance was identified. A full version of this paper is available [3].

Original language	Undefined/Unknown
Pages	180
DOIs	https://doi.org/10.1145/1276958.1276994
Publication status	Published - 2007

Access to Document

10.1145/1276958.1276994

http://doi.acm.org/10.1145/1276958.1276994

Cite this

@conference{fee89e6a53d54720bc5485981424ae56,

title = "Screening the parameters affecting heuristic performance",

abstract = "This research screens the tuning parameters of a combinatorial optimization heuristic. Specifically, it presents a Design of Experiments (DOE) approach that uses a Fractional Factorial Design to screen the tuning parameters of Ant Colony System (ACS) for the Travelling Sales person problem. Screening is a preliminary step towards building a full Response Surface Model (RSM) [2]. It identifies parametersthat have little influence on performance and can be omittedfrom the RSM design. This reduces the complexity andexpense of the RSM design. 10 algorithm parameters and 2 problem characteristics are considered. Open questionson the effect of 3 parameters on performance are answered.A further parameter, sometimes assumed important, was shown to have no effect on performance. A new problem characteristic that effects performance was identified. A full version of this paper is available [3].",

author = "Enda Ridge and Daniel Kudenko",

year = "2007",

doi = "10.1145/1276958.1276994",

language = "Undefined/Unknown",

pages = "180",

}

TY - CONF

T1 - Screening the parameters affecting heuristic performance

AU - Ridge, Enda

AU - Kudenko, Daniel

PY - 2007

Y1 - 2007

N2 - This research screens the tuning parameters of a combinatorial optimization heuristic. Specifically, it presents a Design of Experiments (DOE) approach that uses a Fractional Factorial Design to screen the tuning parameters of Ant Colony System (ACS) for the Travelling Sales person problem. Screening is a preliminary step towards building a full Response Surface Model (RSM) [2]. It identifies parametersthat have little influence on performance and can be omittedfrom the RSM design. This reduces the complexity andexpense of the RSM design. 10 algorithm parameters and 2 problem characteristics are considered. Open questionson the effect of 3 parameters on performance are answered.A further parameter, sometimes assumed important, was shown to have no effect on performance. A new problem characteristic that effects performance was identified. A full version of this paper is available [3].

AB - This research screens the tuning parameters of a combinatorial optimization heuristic. Specifically, it presents a Design of Experiments (DOE) approach that uses a Fractional Factorial Design to screen the tuning parameters of Ant Colony System (ACS) for the Travelling Sales person problem. Screening is a preliminary step towards building a full Response Surface Model (RSM) [2]. It identifies parametersthat have little influence on performance and can be omittedfrom the RSM design. This reduces the complexity andexpense of the RSM design. 10 algorithm parameters and 2 problem characteristics are considered. Open questionson the effect of 3 parameters on performance are answered.A further parameter, sometimes assumed important, was shown to have no effect on performance. A new problem characteristic that effects performance was identified. A full version of this paper is available [3].

U2 - 10.1145/1276958.1276994

DO - 10.1145/1276958.1276994

M3 - Paper

SP - 180

ER -