Kommenda, B. Burlacu, G. Kronberger, M. Affenzeller - Parameter Identification for Symbolic Regression using Nonlinear Least Squares, Genetic Programming and Evolvable Machines, 2019, Springer

Experimental results: Results.csv

GECCO 2019

The Genetic and Evolutionary Computation Conference 13th to 17th July, 2019, Prague, Czech Republic

Parsimony Measures in Multi-objective Genetic Programming for Symbolic Regression

Poster Paper

GECCO_2019_Poster_Burlacu.pdf

Full Paper (not accepted)

GECCO_2019_Burlacu.pdf

Aggregated Results (Excel)

Results_GECCO_2019_Burlacu.7z

2018

Applied Artificial Intelligence Journal

Fechter, A. Beham, S. Wagner and M. Affenzeller - Approximate Q-Learning for Stacking Problems with Continuous Production and Retrieval, Applied Artificial Intelligence, 2018, DOI: 10.1080/08839514.2018.1525852 link

Test Instances:

Instance file format description
Instance 1.1
Instance 1.2
Instance 1.3
Instance 2.1
Instance 2.2
Instance 2.3
Instance 3.1
Instance 3.2
Instance 3.3

Source code implemented in HeuristicLab:

HeuristicLab 3.3 Script

GECCO 2018

The Genetic and Evolutionary Computation Conference 15th to 19th July, 2018, Kyoto, Japan

Algorithm Selection on Generalized Quadratic Assignment Problem Landscapes

Problem Instances

GQAPInstances.zip

Algorithm Selection Data

AlgorithmSelectionData.zip

Schema-based Diversification in Genetic Programming

HeuristicLab Build

heuristiclab.7z

OSGP Experiment

osgp.7z

OSGPS-S Experiments

osgp-s.7z

Applied Soft Computing

Kronberger, M. Kommenda, E. Lughofer, S. Saminger-Platz, A. Promberger, F. Nickel, S. Winkler, M. Affenzeller - Robust Generalized Fuzzy Modeling and Enhanced Symbolic Regression for Modeling Tribological Systems, Applied Soft Computing, 2018 link

Data sets:

README.txt
CF1.csv
CF2.csv
CF3.csv
CF4.csv
NvhRating.csv
Temp1.csv
Temp2.csv
Wear1.csv
Wear2.csv

2017

IJSPM 2017

International Journal of Simulation and Process Modelling

Novel Robustness Measures for Engineering Design Optimisation

Result Tables

Uncertainty Analysis.xlsx
Violation Comparison.xlsx

GECCO 2017

The Genetic and Evolutionary Computation Conference 15th to 19th July, 2017, Berlin, Germany

Instance-based algorithm selection on quadratic assignment problem landscapes

Algorithm Performance and FLA Data

2016

Annals of Operations Research

Vonolfen, M. Affenzeller - Distribution of waiting time for dynamic pickup and delivery problems, Annals of Operations Rsearch, Volume 236, Issue 2, pp 359 - 382, 2018 link

The following additional material is provided:

Test Environment (Executable including Sample)
Source code of the waiting strategies
Benchmark Instances

International Journal of Simulation and Process Modelling

S.M. Winkler, B. Castaño, S. Luengo, S. Schaller, G. Kronberger, M. Affenzeller - Heterogeneous model ensembles for short-term prediction of stock market trends, International Journal of Simulation and Process Modelling, 11(6), pp 504-513, 2016 link

Data set containing preprocessed time series of Spanish stocks: spanish_stock_data_3class.csv

EMSS 2016

28th European Modeling & Simulation Symposium, 26th to 28th September, 2016, Larnaca, Cyprus

Analysis of Uncertainty in Engineering Design Optimization Problems

Problem Implementation

PressureVessel.hl
SpeedReducer.hl
TensionCompressionSpring.hl
WeldedBeam.hl

Solved Problems Experiment

Optimize all Problems Experiment.hl

Result Tables

Solutions.xlsx
Uncertainty Comparison.xlsx

2015

EMSS 2015

27th European Modeling & Simulation Symposium, 21st to 23rd September, 2015, Bergeggi, Italy

Modelling a Clustered Generalized Quadratic Assignment Problem

Problem Instances

10-50-38.txt
10-50-51.txt
10-50-77.txt
15-35-45.txt
15-35-61.txt
15-35-91.txt
20-30-45.txt
20-30-61.txt
20-30-91.txt

2014

IEEE Transactions on Industrial Informatics

Hutterer, A. Beham, M. Affenzeller, F. Auinger and S. Wagner - Software-Enabled Investigation in Metaheuristic Power Grid Optimization, IEEE Transactions on Industrial Informatics, Volume 10, No. 1, pp. 364-372, Feb. 2014 doi: 10.1109/TII.2013.2276525 link

The described experiments shall be made available as follows. For running the simulation model, MatPower is needed with Matlab (link below). The HeuristicLab file contains the optimization runs as well as results that have been analyzed within this paper.

Simulation Model & Optimization Experiments

Simulation File for Matlab Simulation Model
Matpower Data File 118-Bus Data File
HeuristicLab 3.3 Experiment Files HeuristicLab Experiments Collection

Simulation-based optimization with HeuristicLab

A description of how to provide an evaluation service for AnyLogic simulation models is given here.

Links

MATPOWER - A MATLAB Power System Simulation Package, retrieved 11.1.2013
Protocol buffers, retrieved 11.1.2013
Matlab Automation Server, retrieved 11.1.2013

PPSN 2014

13th International Conference on Parallel Problem Solving from Nature, 13th to 17th September, 2014, Ljubljana, Slovenia

Tutorial Algorithm and Experiment Design with HeuristicLab

Tutorial slides

Demo experiment and results: exp.hl exp_results.hl

Symbolic Regression Dataset: ppsn2014.csv

2013

GPTP 2013

Genetic Programming Theory and Practice Workshop, 9th to 11th May, 2013, Ann Arbor, USA

Gaining Deeper Insights in Symbolic Regression: Theoretical and Practical Issues (Keynote)

Slides

Tutorial Algorithm and Experiment Design with HeuristicLab

Tutorial slides

Demo experiment and results: exp.hl exp_results.hl

GECCO 2013

Genetic and Evolutionary Computation Conference, July 6th to 10th, 2013, Amsterdam, Netherlands

Effects of Constant Optimization by Nonlinear Least Squares Minimization in Symbolic Regression

Kommenda, G. Kronberger, S. Winkler, M. Affenzeller and S. Wagner - Effects of constant optimization by nonlinear least squares minimization in symbolic regression, Proceedings of the 15th annual conference companion on Genetic and evolutionary computation, ACM, 2013 link

Slides

Experiments and Results:
Friedman-2.hl
Keijzer-6.hl
Nguyen-7.hl
Pagie-1.hl
Poly-10.hl
Tower.hl
Vladislavleva-4.hl

2012

ECML-PKDD 2012

European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), September 24th to 28th, 2012, Bristol, UK

HeuristicLab Demo: Knowledge Discovery through Symbolic Regression with HeuristicLab

All material is available at the dedicated demo page http://dev.heuristiclab.com/AdditionalMaterial/ECML-PKDD

GECCO 2012

Genetic and Evolutionary Computation Conference, 7th to 11th July, 2012, Philadelphia, USA

Tutorial Algorithm and Experiment Design with HeuristicLab

Tutorial slides

Demo TSP instance from TSPLIB: ch130.tsp ch130.opt.tour

Demo experiment and results: exp.hl exp_results.hl

Demo dataset for symbolic regression: poly-10.csv

Demo dataset for symbolic classification: mammography.csv

APCast 2012

14th International Asia Pacific Conference on Computer Aided System Theory, 6th to 8th February, 2012, Sydney, Australia

Tutorial Algorithm and Experiment Design with HeuristicLab

Tutorial slides

Demo TSP instance from TSPLIB: ch130.tsp ch130.opt.tour

Demo experiment and results: exp.hl exp_results.hl

2011

13th International Conference on Computer Aided Systems Theory (eurocast)

Tutorial Algorithm and Experiment Design with HeuristicLab

Demo dataset for symbolic regression: polynomial.csv

Demo dataset for symbolic time series modeling: Mackey-Glass-17.txt

GECCO 2011

Tutorial Algorithm and Experiment Design with HeuristicLab

Demo TSP instance from TSPLIB: ch130.tsp ch130.opt.tour

Demo experiment and results: exp.hl exp_results.hl

Demo dataset for symbolic regression: poly-10.csv

Demo dataset for symbolic classification: mammography.csv

evo* 2011

Macro-economic Time Series Modeling and Interaction Networks

Data set of macro economic variables: macroeconomicdata.txt

ICCGI 2011

6th International Multi-Conference on Computing in the Global Information Technology, 19th of June, 2011, Luxemburg

Tutorial Algorithm and Experiment Design with HeuristicLab

Demo dataset for symbolic regression: poly-10.csv

Demo dataset for symbolic classification: mammography.csv

LINDI 2011

3rd IEEE International Symposium on Logistics and Industrial Informatics, August 25-27, 2011 in Budapest, Hungary

Demo dataset for warehouse slotting

Stock keeping units: SKUsLindi.txt

Order profile: OrderProfileLindi.txt

IMMM 2011

1st International Conference on Advances in Information Mining and Management, 23th of October, 2011, Barcelona

System Identification and Data Mining with HeuristicLab

Demo TSP instance from TSPLIB: ch130.tsp ch130.opt.tour

Demo experiment and results: exp.hl exp_results.hl

Demo dataset for symbolic regression: poly-10.csv

Demo dataset for symbolic classification: mammography.csv

2010

22nd European Modeling & Simulation Symposium (EMSS)

Mutation Effects in Genetic Algorithms with Offspring Selection Applied to Combinatorial Optimization Problems

Authors: S. Wagner, M. Affenzeller, A. Beham, G. Kronberger, S.M. Winkler

The HeuristicLab experiments described in the paper can be downloaded here.

Dissertation Kronberger

The following datasets are used in experiments in the thesis.

Artificial benchmark datasets

Friedman-I

friedman-I.csv

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).

Variables x01,..., x10 are sampled uniformly from the unit hypercube (x~U(0,1)). Epsilon is generated from the standard normal distribution (e~N(0,1)).

Friedman-II

friedman-II.csv

This dataset is also described in (Friedman, 1991). The signal-to-noise ratio in this dataset is larger compared to the Friedman-I function.

Variables x1,..., x5 are sampled uniformly from the unit hypercube (x~U(0,1)).

Breiman-I

breiman-I.csv

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.

Epsilon is generated from the normal distribution (e~N(0,2)).

Variables x01,..., x10 are randomly sampled attributes following the probability distributions:

Real-world datasets

Chemical-I

chemical-I.csv

Chemical-II

chemical-II.csv

Financial-I

financial-I.csv

Macro-Economic

macro-economic.csv

Housing

housing.csv

References

Jerome H. Friedman, Multivariate adaptive regression splines, The Annals of Statistics, 19(1):1-141, 1991.
Leo Breiman, Jerome H. Friedman, Charles J. Stone and R. A. Olson, Classification and Regression Trees, Chapman and Hall, 1984