A Gender-Based Genetic Algorithm for the Automatic Configuration of Algorithms

By C. Ansótegui, Meinolf Sellmann, Kevin Tierney. International Conference on Principles and Practice of Constraint Programming, 2009

370 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Industrial and Manufacturing Engineering

View Paper: https://www.semanticscholar.org/paper/997dd6c2d49f6d89a4f9b28729163dc18549f3d3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

In Industrial and Manufacturing Engineering

Citations in the corpus

LEAPSANDBOUNDS: A Method for Approximately Optimal Algorithm Configuration (2018, methodology): “Furthermore, they show that the gap between worst-case runtimes of existing algorithms (SMAC, ROAR, ParamILS, GGA, irace) and their solution can be arbitrarily large.”
Evolutionary Bilevel Optimization for Complex Control Tasks (2015, methodology): “To address this issue, GGA [25] divided the upper level population into two genders and only evaluated a subset of the population every generation.”
ISAC - Instance-Specific Algorithm Configuration (2010, methodology): “Furthermore, since it was first released, the default parameters of the solver have been drastically improved by general purpose parameter tuners [2, 17].”
Solving Mixed Integer Programs Using Neural Networks (2020, background): “For example, Hutter et al. (2011) proposes Sequential Model-based Algorithm Configuration (SMAC), which iterates over three main steps: 1) evaluate the performance of selected solver parameter values on a set of instances, 2) learn a model to predict the performance of given parameter values from th”
Automatically improving the anytime behaviour of optimisation algorithms (2014, background): “The natural evolution of this trend has been to tackle algorithm configuration as an optimisation problem [4, 9, 10, 31, 32, 49].”
The irace package: Iterated racing for automatic algorithm configuration (2016, methodology): “The methods used in this context include experimental design techniques [2,29] , racing approaches [20] , and algorithmic methods for parameter conﬁguration, such as heuristic search techniques [3,10,41,73,81] , and statistical modeling approaches [11,43] .”
Parallel Algorithm Configuration (2012, background): “Indeed, the literature on algorithm configuration already contains examples of both the first [1, 4, 2, 3] and second [4, 5] forms of parallelization.”
SMAC3: A Versatile Bayesian Optimization Package for Hyperparameter Optimization (2021, background): “A more general view on the problem of HPO is called algorithm conﬁguration (AC) (Hutter et al., 2009; Ans´otegui et al., 2009; L´opez-Ib´a˜nez et al., 2016), where the goal is to determine a well-performing robust conﬁguration across a set of problem instances i from a ﬁnite set I : The origin of…”
Algorithm Selection for Combinatorial Search Problems: A Survey (2012, methodology): “ISAC (Kadioglu et al., 2010) uses GGA (Ans´otegui, Sellmann, & Tierney, 2009) to automatically tune algorithms for clusters of problem instances.”
Automated Algorithm Configuration and Parameter Tuning (2012, methodology): “Ansótegui et al (2009) recently developed a gender-based genetic algorithm for solving algorithm configuration problems.”
Sequential Model-Based Optimization for General Algorithm Configuration (2011, methodology): “We note that in a previous comparison [9] of GGA and FOCUSEDILS, capping was disabled in FOCUSEDILS; this explains its poor performance there and its better performance here.”
Automated Configuration of Mixed Integer Programming Solvers (2010, neutral): “Some relatively recent model-free approaches permit both larger numbers of parameters and categorical domains, in particular Composer [12], F-Race [9, 8], GGA [3], and our own ParamILS [20, 19].”
An Efficient Approach for Assessing Hyperparameter Importance (2014, methodology): “…optimization methods in that community are based on heuristic search (Adenso-Diaz and Laguna, 2006; Nannen and Eiben, 2007; Hutter et al., 2009; Ansotegui et al., 2009) and racing algorithms (Maron and Moore, 1994; Birattari et al., 2010), recently, Bayesian optimization methods based on…”
Instance-Specific Algorithm Configuration (2014, methodology): “For default Hegel, an average test instance takes 2.10 times the time of ISAC-Hegel while GGA-Hegel only runs 10 % slower.”