Please use this identifier to cite or link to this item: http://hdl.handle.net/11667/74
Appears in Collections:University of Stirling Research Data
Title: Data sets for the paper "Mining Markov Network Surrogates for Value-Added Optimisation" presented at the Genetic and Evolutionary Computation Conference (GECCO) 2016
Creator(s): Brownlee, Alexander
Contact Email: sbr@cs.stir.ac.uk
Keywords: metaheuristics
surrogates
fitness approximation
decision making
Date Available: 10-May-2016
Citation: Brownlee, A (2016): Data sets for the paper "Mining Markov Network Surrogates for Value-Added Optimisation" presented at the Genetic and Evolutionary Computation Conference (GECCO) 2016. University of Stirling. Dataset. http://hdl.handle.net/11667/74
Dataset Description (Abstract): Model coefficients for the models that are analysed in the paper, and predicted/true fitness values for the validation data. As described in the paper: Surrogate fitness functions are a popular technique for speeding up metaheuristics, replacing calls to a costly fitness function with calls to a cheap model. However, surrogates also represent an explicit model of the fitness function, which can be exploited beyond approximating the fitness of solutions. The paper proposed that mining surrogate fitness models can yield useful additional information on the problem to the decision maker, adding value to the optimisation process. An existing fitness model based on Markov networks was presented and applied to the optimisation of glazing on a building facade. Analysis of the model revealed how its parameters point towards the global optima of the problem after only part of the optimisation run, and revealed useful properties like the relative sensitivities of the problem variables.
Dataset Description (TOC): Documentation in README.txt: Data sets for the paper "Mining Markov Network Surrogates for Value-Added Optimisation" by Alexander E.I. Brownlee sbr@cs.stir.ac.uk www.cs.stir.ac.uk/~sbr Presented in the SAEOpt Workshop at GECCO 2016 DOI:10.1145/2908961.2931711 Files: 10runs-cost-biuni-1000-sel400.txt 10runs-cost-uni-400-sel140.txt 10runs-energy-biuni-1000-sel400.txt 10runs-energy-uni-400-sel140.txt These are the aggregated model coefficients for the four MFMs mentione in the paper (cost/energy univariate/univariate+bivariate). Columns are: coeffNumber mean stdDev biuni-cost-predicts.txt biuni-energy-predicts.txt uni-cost-predicts.txt uni-energy-predicts.txt These are the predicted and true fitness values for the test solutions used to compute r^2 for the four MFMs mentione in the paper (cost/energy univariate/univariate+bivariate). coefficients.xlsx These are the model coefficients for the ten repeat runs of each model. Each tab's rows are: rowNumber 120 columns for univariate alphas 240 columns for bivariate alphas if applicable 1 colum for constant
Type: dataset
Contract/Grant Title: DAASE: Dynamic Adaptive Automated Software Engineering
FAIME: A Feature based Framework to Automatically Integrate and Improve Metaheuristics via Examples
Funder(s): EPSRC - Engineering and Physical Sciences Research Council
Contract/Grant Number: EP/J017515/1
EP/N002849/1
URI: http://hdl.handle.net/11667/74
Rights: Rights covered by the standard CC-BY 4.0 licence: https://creativecommons.org/licenses/by/4.0/
Affiliation(s) of Dataset Creator(s): University of Stirling



This item is protected by original copyright



Items in DataSTORRE are protected by copyright, with all rights reserved, unless otherwise indicated.