All workshops will take place on Saturday, September 1st, 2012.
Workshop proposals submission is now closed.
Below you may find a list of accepted workshops with short abstracts. For further information about a particular workshop, please visit the webpage of the workshop.
We would be grateful if you could register to the workshops you intend to attend. The registration is not binding, and your appearance is still optional, but thanks to this information we may be able to plan the workshop schedule appropriately.
Evolving Predictive Systems
Bogdan Gabrys and Athanasios Tsakonas, Bournemouth University, UK
Abstract: In recent years, the data mining scientific community witnessed a very strong demand for predictive systems that will be able to evolve and adapt. The range of tasks fulfilled by evolving predictive systems is very broad and covering many different application areas. Despite the high number of publications dealing with applications, there are still unaddressed pressing issues of evolving predictive systems design and development, such as complexity analysis, ensemble architectures and meta-learning. This session is devoted to the discussion of robust, context aware and easy-to-use evolving predictive systems, which improve, adapt and possibly maintain themselves within their respective environments and constraints. The workshop addresses people from the scientific IT community who are active in the research domain of data-driven systems capable to adapt to changing situations and environments. The considered approaches can include evolutionary algorithms, other nature-inspired methods or heuristic approaches. Contributions presenting research dealing with ensemble architectures, complexity issues (size, form and interpretation of the solution formula, time and algorithm complexity) and meta-learning incorporation are particularly welcome.
Joint Workshop on Automated Selection and Tuning of Algorithms
Part A: Continuous Search Spaces — Focus on Algorithm Selection
Heike Trautmann, Mike Preuss, Olaf Mersmann, and Bernd Bischl, University of Dortmund, Germany
Part B: Discrete Search Spaces — Focus on Parameter Selection
Andrew Parkes and Ender Ozcan, University of Nottingham, UK
Abstract: The steadily growing supply of new optimization methods makes the algorithm selection problem an increasingly pressing and challenging task, both in continuous as well as in discrete combinatorial optimization. Therefore, choosing and tuning a suitable optimization algorithm for a given instance of an optimization problem is a crucial issue and should be supported by automated tools based on problem characteristics.
The aim of this workshop is to collect research concepts and bring together researchers of interdisciplinary areas such as computer science, artificial intelligence, statistics, machine learning and optimization in order to interactively discuss the current state-of-the art and most important research topics to be addressed in the near future.
Theoretical Aspects of Evolutionary Multiobjective Optimization: Interactive Problem Solving Sessions and New Results
Dimo Brockhoff, INRIA, France
Gunter Rudolph, University of Dortmund, Germany
Abstract: Evolutionary Multiobjective Optimization (EMO), i.e., the simultaneous optimization of 2 or more objective functions by means of bio-inspired search heuristics, has become one of the main research fields in evolutionary computation in recent years and as such also gained interest from the classical field of multicriteria decision making. Together with the rapid development in practice, also theoretical analyses of EMO gained more and more interest recently.
To foster this growing interest, the PPSN 2012 workshop on Theoretical Aspects of EMO aims at bringing together both theoreticians working in the field of EMO and from single-objective optimization and EMO practitioners. Besides presenting the newest theoretical results about EMO, the workshop aims at providing hands-on sessions in which practitioners and theoreticians work in collaboration on selected open questions proposed by the audience.
In this regard, we would like to invite both theoreticians and practitioners to either present an open problem or discuss about the latest developments in the field of theoretical analysis of evolutionary multiobjective optimization.
Modeling Biological Systems Workshop
Julia Handl and Joshua Knowles, University of Manchester, UK
Yaochu Jin, University of Surrey, UK
Abstract: The aim of the workshop is to provide a dedicated forum for discussion amongst those who are interested in the modeling of biological systems at various scales, ranging from molecular systems to ecological and social interactions. We thus invite papers on all aspects of computational modeling, including applied optimization and simulated evolution, and approaches founded in complexity theory and artificial life.
Topics of interest include, but are not limited to: network biology, signalling pathway modelling, evolution of complexity or "complexification", adaptive walks and fitness landscapes, the benefit of sex, non-adaptive evolution, drift and neutrality, reconstruction of gene regulatory networks, protein folding, structure prediction and docking modularity, robustness and evolvability in biological systems, self-organization and emergence, evolutionary game theory, immune systems modeling, embryogeny, development and pattern formation, models for neural plasticity and neural development, artificial development, and biomimetics.
Parallel Techniques in Search, Optimization, and Learning
Enrique Alba and Francisco Luna, University of Málaga, Spain
Abstract: Modern research during these last twenty years has expanded to address very interesting problems of large complexity (dimensionality, restrictions, computing intensive...). In particular, those coming from real-world scenarios are getting both larger in size and harder in complexity.
Aiming at finding accurate (and robust) solutions in the shortest possible computational time, these problems face researchers to new challenges of difficult solution with traditional techniques and computers. One way to achieve unseen numerical and efficient results is the use of parallel algorithms, hardware, and specialized techniques.
With the evolution of parallel architectures (symmetric multiprocessors, multi/many-cores, GPUs, etc.), many opportunities emerge for the design of efficient algorithms.
This workshop seeks contributions on new theoretical advances and carefully designed, well-analyzed proposals in the field of parallel search algorithms. It is also intended to gather researchers from several domains (operations research, computer science, management science, communications and networks, ...) with an opportunity for presenting and discussing their more recent developments in theory and application of parallel search algorithms. An open atmosphere for discussion of future research lines will hopefully help in defining where we are and where are we going in this crossroad between parallelism and (Nature) problem solving.