Mark Wineberg

Headshot of Mark Wineberg
Associate Professor
School of Computer Science
Email: 
mwineber@uoguelph.ca
Office: 
REYN 2202
Website links: 
GitHub

Find Related People by Keyword

Artificial Intelligence, Artificial life, Computational intelligence, Evolutionary Computation

Education and Employment Background

Dr. Mark Wineberg received his PhD from Carleton University in 2000. He joined the School of Computer Science at the University of Guelph in 2000 where he is now an Associate Professor.

Research Themes

In his research, Wineberg examines and models the behaviour of computational evolutionary algorithms (EA) in complex environments that benefit from advanced structures within the EA. He explores the beahviour of EAs with multi-populations, such as co-evolutionary systems and hierarchical systems. He develops representational diagrams to design and reason about such complex systems and advanced statistical analysis for comparing their beahviour. Key areas of focus include:

  1. Component hierarchies. These novel hierarchical EA systems are coevolutionary systems with many interacting populations. Wineberg created representational diagrams to design and reason about multipopulational systems.
  2. Multiobjective optimization. Using evolutionary algorithms, Wineberg examines multiobjective optimization. He explores the idea of robustness as an important general non-functional requirement. Robust solutions are those that when variation occurs in the implementation of the solution, the functional behaviour of the solution does not drastically change.
  3. Analyzing sampling patterns in the search space. Wineberg and his team compared and contrasted how two current state-of-the-art algorithms for real-valued genomes search through the fitness landscape.
  4. Performing statistical analysis of experiments on EA systems. Wineberg evaluates statistical analysis practices used by researchers and seeks to develop analysis techniques that are tailored to the observed properties of EC systems. Doing so will enable the application of more advanced statistical models to EC experimental results.

Highlights

  • Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery grant, 2001-2011
  • Conference Program Committee Member, Congress on Evolutionary Computation, 2006, 2007
  • Editor, Genetic Programming and Evolvable Machines Journal, 2005