Changes between Version 45 and Version 46 of AdditionalMaterial
- Timestamp:
- 06/28/12 10:47:19 (12 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
AdditionalMaterial
v45 v46 3 3 This page contains a collection of additional material related to publications of members of the research group HEAL. 4 4 5 == 201 0==5 == 2012 == 6 6 7 === 22nd European Modeling & Simulation Symposium (EMSS) === 7 === GECCO 2012 === 8 Genetic and Evolutionary Computation Conference, 9 7th to 11th July, 2012, Philadelphia, USA 8 10 9 ==== Mutation Effects in Genetic Algorithms with Offspring Selection Applied to Combinatorial Optimization Problems ==== 11 ==== Tutorial Algorithm and Experiment Design with HeuristicLab ==== 12 [export:/misc/publications/2012/conferences/GECCO/swagner/slides.pdf Tutorial slides] 10 13 11 Authors: S. Wagner, M. Affenzeller, A. Beham, G. Kronberger, S.M. Winkler 14 Demo TSP instance from [http://comopt.ifi.uni-heidelberg.de/software/TSPLIB95 TSPLIB]. 15 [export:/misc/publications/2012/conferences/GECCO/swagner/ch130.tsp ch130.tsp] 16 [export:/misc/publications/2012/conferences/GECCO/swagner/ch130.opt.tour ch130.opt.tour] 12 17 13 The !HeuristicLab experiments described in the paper can be downloaded [source:misc/publications/2010/conferences/EMSS/swagner here]. 18 Demo experiment and results. 19 [export:/misc/publications/2012/conferences/GECCO/swagner/exp.hl exp.hl] 20 [export:/misc/publications/2012/conferences/GECCO/swagner/exp_results.hl exp_results.hl] 14 21 15 === Dissertation Kronberger === 16 The following datasets are used in experiments in the thesis. 17 ==== Artificial benchmark datasets ==== 18 ===== Friedman-I ===== 19 [export:/misc/publications/2010/theses/gkronber/friedman-I.csv friedman-I.csv] 22 Demo dataset for symbolic regression. 23 [export:/misc/publications/2012/conferences/GECCO/gkronber/poly-10.csv poly-10.csv] 20 24 21 This dataset is described in (Friedman, 1991), where it is used to benchmark the multi-variate adaptive regression splines (MARS) algorithm. The signal-to-noise ratio in this dataset is rather low, so it is difficult to rediscover the generating function f(x) especially the terms below the noise level (x4 and x5). 25 Demo dataset for symbolic classification. 26 [export:/misc/publications/2012/conferences/GECCO/gkronber/mammography.csv mammography.csv] 22 27 23 [[Image(friedman-I.png)]] 28 === IEEE Transactions on Industrial Informatics === 29 Special section on ''Software Engineering in Factory and Energy Automation'' 24 30 25 Variables x01,..., x10 are sampled uniformly from the unit hypercube (x~U(0,1)). 26 Epsilon is generated from the standard normal distribution (e~N(0,1)).31 ==== Simulation-based optimization with HeuristicLab ==== 32 A description of how to provide an evaluation service for [http://www.xjtek.com/ AnyLogic] simulation models is given [wiki:UsersHowtosOptimizingAnyLogicModels here]. 27 33 28 ==== = Friedman-II =====29 [export:/misc/publications/2010/theses/gkronber/friedman-II.csv friedman-II.csv]34 ==== Links ==== 35 [http://www.interpss.org/ InterPSS] - Internet technology based Power System Simulator, retrieved 7.12.2011 30 36 31 This dataset is also described in (Friedman, 1991). The signal-to-noise ratio in this dataset is larger compared to the Friedman-I function. 37 [http://www.pserc.cornell.edu/matpower/ MATPOWER] - A MATLAB Power System Simulation Package, retrieved 7.12.2011 32 38 33 [[Image(friedman-ii.png)]] 39 [http://code.google.com/apis/protocolbuffers/docs/overview.html Protocol buffers], retrieved 11.12.2011 34 40 35 Variables x1,..., x5 are sampled uniformly from the unit hypercube (x~U(0,1)). 41 === APCast 2012 === 42 14th International Asia Pacific Conference on Computer Aided System Theory, 43 6th to 8th February, 2012, Sydney, Australia 36 44 37 ===== Breiman-I ===== 38 [export:/misc/publications/2010/theses/gkronber/breiman-I.csv breiman-I.csv] 39 This dataset is described in (Breiman et al., 1984), where it is used to benchmark the classification and regression trees (CART) algorithm. The signal-to-noise ratio is rather low and additionally it contains a crisp conditional which makes it rather difficult to rediscover the generating function with a symbolic regression approach. 45 ==== Tutorial Algorithm and Experiment Design with HeuristicLab ==== 46 [export:/misc/publications/2012/conferences/APCast/swagner/slides.pdf Tutorial slides] 40 47 41 [[Image(breiman-I.png)]] 48 Demo TSP instance from [http://comopt.ifi.uni-heidelberg.de/software/TSPLIB95 TSPLIB]. 49 [export:/misc/publications/2012/conferences/APCast/swagner/ch130.tsp ch130.tsp] 50 [export:/misc/publications/2012/conferences/APCast/swagner/ch130.opt.tour ch130.opt.tour] 42 51 43 Epsilon is generated from the normal distribution (e~N(0,2)). 44 45 Variables x01,..., x10 are randomly sampled attributes following the probability distributions: 46 47 [[Image(breiman-I-variables.png)]] 48 49 ==== Real-world datasets ==== 50 51 ===== Chemical-I ===== 52 [export:/misc/publications/2010/theses/gkronber/chemical-I.csv chemical-I.csv] 53 54 ===== Chemical-II ===== 55 [export:/misc/publications/2010/theses/gkronber/chemical-II.csv chemical-II.csv] 56 57 ===== Financial-I ===== 58 [export:/misc/publications/2010/theses/gkronber/financial-I.csv financial-I.csv] 59 60 ===== Macro-Economic ===== 61 [export:/misc/publications/2010/theses/gkronber/macro-economic.csv macro-economic.csv] 62 63 ===== Housing ===== 64 [export:/misc/publications/2010/theses/gkronber/housing.csv housing.csv] 65 66 ==== References ==== 67 Jerome H. Friedman, Multivariate adaptive regression splines, ''The Annals of Statistics'', 19(1):1-141, 1991. \\ 68 Leo Breiman, Jerome H. Friedman, Charles J. Stone and R. A. Olson, ''Classification and Regression Trees'', Chapman and Hall, 1984\\ 52 Demo experiment and results. 53 [export:/misc/publications/2012/conferences/APCast/swagner/exp.hl exp.hl] 54 [export:/misc/publications/2012/conferences/APCast/swagner/exp_results.hl exp_results.hl] 69 55 70 56 ---- … … 145 131 ---- 146 132 147 == 2012 ==148 133 149 === IEEE Transactions on Industrial Informatics === 150 Special section on ''Software Engineering in Factory and Energy Automation'' 134 == 2010 == 151 135 152 ==== Simulation-based optimization with HeuristicLab ==== 153 A description of how to provide an evaluation service for [http://www.xjtek.com/ AnyLogic] simulation models is given [wiki:UsersHowtosOptimizingAnyLogicModels here]. 136 === 22nd European Modeling & Simulation Symposium (EMSS) === 154 137 155 ==== Links ==== 156 [http://www.interpss.org/ InterPSS] - Internet technology based Power System Simulator, retrieved 7.12.2011 138 ==== Mutation Effects in Genetic Algorithms with Offspring Selection Applied to Combinatorial Optimization Problems ==== 157 139 158 [http://www.pserc.cornell.edu/matpower/ MATPOWER] - A MATLAB Power System Simulation Package, retrieved 7.12.2011 140 Authors: S. Wagner, M. Affenzeller, A. Beham, G. Kronberger, S.M. Winkler 159 141 160 [http://code.google.com/apis/protocolbuffers/docs/overview.html Protocol buffers], retrieved 11.12.2011 142 The !HeuristicLab experiments described in the paper can be downloaded [source:misc/publications/2010/conferences/EMSS/swagner here]. 161 143 162 === APCast 2012 === 163 14th International Asia Pacific Conference on Computer Aided System Theory, 164 6th to 8th February, 2012, Sydney, Australia 144 === Dissertation Kronberger === 145 The following datasets are used in experiments in the thesis. 146 ==== Artificial benchmark datasets ==== 147 ===== Friedman-I ===== 148 [export:/misc/publications/2010/theses/gkronber/friedman-I.csv friedman-I.csv] 165 149 166 ==== Tutorial Algorithm and Experiment Design with HeuristicLab ==== 167 [export:/misc/publications/2012/conferences/APCast/swagner/slides.pdf Tutorial slides] 150 This dataset is described in (Friedman, 1991), where it is used to benchmark the multi-variate adaptive regression splines (MARS) algorithm. The signal-to-noise ratio in this dataset is rather low, so it is difficult to rediscover the generating function f(x) especially the terms below the noise level (x4 and x5). 168 151 169 Demo TSP instance from [http://comopt.ifi.uni-heidelberg.de/software/TSPLIB95 TSPLIB]. 170 [export:/misc/publications/2012/conferences/APCast/swagner/ch130.tsp ch130.tsp] 171 [export:/misc/publications/2012/conferences/APCast/swagner/ch130.opt.tour ch130.opt.tour] 152 [[Image(friedman-I.png)]] 172 153 173 Demo experiment and results. 174 [export:/misc/publications/2012/conferences/APCast/swagner/exp.hl exp.hl] 175 [export:/misc/publications/2012/conferences/APCast/swagner/exp_results.hl exp_results.hl] 154 Variables x01,..., x10 are sampled uniformly from the unit hypercube (x~U(0,1)). 155 Epsilon is generated from the standard normal distribution (e~N(0,1)). 176 156 177 === GECCO 2012 === 178 Genetic and Evolutionary Computation Conference, 179 7th to 11th July, 2012, Philadelphia, USA 157 ===== Friedman-II ===== 158 [export:/misc/publications/2010/theses/gkronber/friedman-II.csv friedman-II.csv] 180 159 181 ==== Tutorial Algorithm and Experiment Design with HeuristicLab ==== 182 [export:/misc/publications/2012/conferences/GECCO/swagner/slides.pdf Tutorial slides] 160 This dataset is also described in (Friedman, 1991). The signal-to-noise ratio in this dataset is larger compared to the Friedman-I function. 183 161 184 Demo TSP instance from [http://comopt.ifi.uni-heidelberg.de/software/TSPLIB95 TSPLIB]. 185 [export:/misc/publications/2012/conferences/GECCO/swagner/ch130.tsp ch130.tsp] 186 [export:/misc/publications/2012/conferences/GECCO/swagner/ch130.opt.tour ch130.opt.tour] 162 [[Image(friedman-ii.png)]] 187 163 188 Demo experiment and results. 189 [export:/misc/publications/2012/conferences/GECCO/swagner/exp.hl exp.hl] 190 [export:/misc/publications/2012/conferences/GECCO/swagner/exp_results.hl exp_results.hl] 164 Variables x1,..., x5 are sampled uniformly from the unit hypercube (x~U(0,1)). 191 165 192 Demo dataset for symbolic regression. 193 [export:/misc/publications/2012/conferences/GECCO/gkronber/poly-10.csv poly-10.csv] 166 ===== Breiman-I ===== 167 [export:/misc/publications/2010/theses/gkronber/breiman-I.csv breiman-I.csv] 168 This dataset is described in (Breiman et al., 1984), where it is used to benchmark the classification and regression trees (CART) algorithm. The signal-to-noise ratio is rather low and additionally it contains a crisp conditional which makes it rather difficult to rediscover the generating function with a symbolic regression approach. 194 169 195 Demo dataset for symbolic classification. 196 [export:/misc/publications/2012/conferences/GECCO/gkronber/mammography.csv mammography.csv] 170 [[Image(breiman-I.png)]] 171 172 Epsilon is generated from the normal distribution (e~N(0,2)). 173 174 Variables x01,..., x10 are randomly sampled attributes following the probability distributions: 175 176 [[Image(breiman-I-variables.png)]] 177 178 ==== Real-world datasets ==== 179 180 ===== Chemical-I ===== 181 [export:/misc/publications/2010/theses/gkronber/chemical-I.csv chemical-I.csv] 182 183 ===== Chemical-II ===== 184 [export:/misc/publications/2010/theses/gkronber/chemical-II.csv chemical-II.csv] 185 186 ===== Financial-I ===== 187 [export:/misc/publications/2010/theses/gkronber/financial-I.csv financial-I.csv] 188 189 ===== Macro-Economic ===== 190 [export:/misc/publications/2010/theses/gkronber/macro-economic.csv macro-economic.csv] 191 192 ===== Housing ===== 193 [export:/misc/publications/2010/theses/gkronber/housing.csv housing.csv] 194 195 ==== References ==== 196 Jerome H. Friedman, Multivariate adaptive regression splines, ''The Annals of Statistics'', 19(1):1-141, 1991. \\ 197 Leo Breiman, Jerome H. Friedman, Charles J. Stone and R. A. Olson, ''Classification and Regression Trees'', Chapman and Hall, 1984\\ 198