Funded PhD Position: Exact Tracing of Evolutionary Search Trajectories in Complex Hypothesis Spaces

The Heuristic and Evolutionary Algorithms Laboratory at the Upper Austria University of Applied Sciences (Austria) offers a funded PhD position within the scope of the International PhD Program in Informatics Hagenberg. The group conducts scientific research in the field of metaheuristics for solving business, engineering, and biomedical problems.

The focus of the PhD research project should be to gain a better understanding of the evolutionary process by analyzing the search trajectories of genetic programming (GP) algorithms in different hypothesis spaces for academic as well as real-world data mining applications.

Recent algorithmic achievements like offspring selection in combination with linear scaling have enabled GP to achieve high quality results in system identification in less than 50 generations using populations with only several hundred individuals. Therefore the active gene pool of evolutionary search remains manageable and may be subjected to new theoretical investigations closely related to genetic programming schema theories, building block hypotheses and bloat theories.

We are seeking highly motivated and creative students with a master degree in Mathematics/Informatics or a related technical field. Prior knowledge about heuristic optimization and machine learning and programming experience with C# is a plus. Applications (including your CV, degree certificates, list of publications etc.) and informal queries about the lab and research projects should be addressed to Dr. Michael Affenzeller (, phone +43 (0)7236 3888-2031).

Further details about the PhD program can be found on

A longer, printable description of the PhD position as PDF document is also available.

Application Deadline: July 15th, 2011.

HeuristicLab 3.3.4 Released

The HeuristicLab development team is pleased to announce that you can now download HeuristicLab 3.3.4. We have added some exciting new features in this last release, such as:
  • New algorithms: Particle Swarm Optimization, Variable Neighborhood Search, Support Vector Machines for Classification
  • New problems: Quadratic Assignment Problem, Multi-objective Symbolic Regression Problem

For the full feature list, samples and documentation, go to

FFH 2011: Best Poster Award Runner-Up

Michael Affenzeller and Stefan Wagner represented HEAL at the 5th Austrian FH Research Forum (FFH 2011), which took place in Vienna from 27. to 28. April 2011. The aim of the event was to highlight recent scientific research conducted at Universities of Applied Sciences and network with fellow researchers, decision makers and business representatives. In total, 165 papers and posters from diverse research branches were presented at the forum. Our contribution, "HeuristicLab 3.3: An Extensible and Flexible Software System
for Heuristic Optimization", received the runner-up best poster award.

You can now download the poster.

Meet HEAL team members at GECCO 2011

We are very pleased to announce the following HEAL team members have been accepted to present recent research at GECCO 2011 in Dublin, Ireland from 12-16 July 2011.
  • HeuristicLab head developer Stefan Wagner and Gabriel Kronberger will demonstrate how to apply and analyze metaheuristics using our open source optimization environment HeuristicLab. More information is available on the HeuristicLab Tutorial homepage.
  • In addition, Gabriel Kronberger will give a talk about "Overfitting Detection and Adaptive Covariant Parsimony Pressure for Symbolic Regression" in the Symbolic Regression and Modeling Workshop.
  • In the Genetic Programming Track, Stephan Winkler will present a poster on the "Analysis of the Effects of Enhanced Selection Concepts for Genetic Programming Based Structure Identification Using Fine-Grained Population Diversity Estimation".
  • You can also meet Stephan in the Workshop on Medical Applications of Genetic and Evolutionary Computation and listen to his talk about the "Identification of Cancer Diagnosis Estimation Models Using Evolutionary Algorithms - A Case Study for Breast Cancer, Melanoma, and Cancer in the Respiratory System", a topic he has been working on in close cooperation with the General Hospital Linz (AKh Linz).
  • Finally, Viktoria Dorfer will give a talk "On the Performance of Evolutionary Algorithms in Biomedical Keyword Clustering", also in the Workshop on Medical Applications of Genetic and Evolutionary Computation.
Michael Affenzeller and the HEAL team are looking forward to GECCO 2011 and hope to see you there!

evostar 2011

Next week, HEAL researchers will be attending the evo* conference in Torino, Italy. evo* consists of four co-located conferences in the field of evolutionary computing, namely eurogp, evobio, evocop and evoapplications. Together, they form the premier multi-conference on evolutionary computing in Europe.

  • Michael Kommenda will present his work on "Data Mining using Unguided Symbolic Regression on a Blast Furnace Dataset" in the evonum workshop on on bio-inspired algorithms for continuous parameter optimisation.
  • Gabriel Kronberger will present results on "Macro-economic Time Series Modeling and Interaction Networks" in the evofin workshop on applications in finance and economics.


Subscribe to News