A Racing Algorithm for Configuring Metaheuristics

By Mauro Birattari, Thomas Stützle, Luís Paquete, Klaus Varrentrapp. 2002

647 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Efficient Benchmarking of Hyperparameter Optimizers via Surrogates (2015, methodology): “In fact, several researchers have already applied this approach to simplify experiments (Bardenet et al. 2013; Snoek, Larochelle, and Adams 2012; Birattari et al. 2002).”
ParamILS: An Automatic Algorithm Configuration Framework (2014, methodology): “Given a training set consisting of M problem instances, for N ≤ M , we drew a sample of N instances uniformly at random and without replacement and added them to the list.”
LEAPSANDBOUNDS: A Method for Approximately Optimal Algorithm Configuration (2018, methodology): “, 2011; 2013), irace (Birattari et al., 2002; López-Ibánez et al., 2011), and GGA (Ansótegui et al.”
Parallel metaheuristics: recent advances and new trends (2012, background): “RACE (Birattari et al., 2002) and SPO (Bartz-Beielstein and Preuss, 2009) are interesting approaches for validating and reducing the effort of the researcher in experimentation.”
Automatically improving the anytime behaviour of optimisation algorithms (2014, methodology): “F-Race [12] is a racing procedure [46] for the selection of the best among a given set of algorithm configurations, by means of the non-parametric Friedman’s two-way analysis of variance by ranks, and its associated post-hoc test [16].”
Boosting Verification by Automatic Tuning of Decision Procedures (2007, background): “Very broadly, automated parame ter optimization can be seen as as a stochastic optimization problem that can be solved using a range of generic and specific methods [13], [14], [15].”
Automated configuration of algorithms for solving hard computational problems (2009, methodology): “In particular, we would like to compare the methods studied in this thesis to racing algorithms, such as F-Race (Birattari et al., 2002; Birattari, 2004) and iterated F-Race (Balaprakash et al.”
Learned lessons in credit card fraud detection from a practitioner perspective (2014, methodology): “Dal Pozzolo et all [48] recently proposed to use a F-race [49] algorithm to automatically select the correct unbalanced strategy for a given dataset.”
Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learning (2020, methodology): “This problem setup can also be used to introduce generalizations of the algorithm selection problem such as algorithm configuration (Birattari et al., 2002; Hutter et al., 2009; Kleinberg et al., 2017), per-instance algorithm configuration (Xu et al., 2010; Malitsky et al., 2012) and dynamic…”
ISAC - Instance-Specific Algorithm Configuration (2010, background): “For a small set of possible parameter configurations, [3] employs a racing mechanism.”
The irace package: Iterated racing for automatic algorithm configuration (2016, methodology): “The F-Race algorithm [6, 7] relies on the non-parametric Friedman’s two-way analysis of variance by ranks, the Friedman test, and its associated post-hoc test, as described by Conover [9].”
Hyperparameter optimization: Foundations, algorithms, best practices, and open challenges (2021, methodology): “The statistical test that discards individuals can be the Friedman test or the t-test (Birattari et al., 2002), the latter possibly with multiple testing correction.”
Automated Configuration of Mixed Integer Programming Solvers (2010, unknown): “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].”
Sequential Model-Based Optimization for General Algorithm Configuration (2011, methodology): “F-RACE and its extensions have been used to optimize various high-performance algorithms, including iterated local search and ant colony optimization procedures for timetabling tasks and the travelling salesperson problem [6, 5].”
A Survey of Automatic Parameter Tuning Methods for Metaheuristics (2020, methodology): “F-Race, which is inspired from the algorithm Hoeffding race [49], [50] in machine learning for model selection, was proposed in [45] and comprehensively studied in [31].”
A Gender-Based Genetic Algorithm for the Automatic Configuration of Algorithms (2009, methodology): “In our setting, we could employ the racing framework from [2] for selecting the top competitive individuals.”