schedule

Graduate Student Workshop
Michael O'Neill

[Summary] [Further details]

Dave Davis

Industrial participants who would like to interact with the consultants during this workshop are encouraged to send email in advance to describing the type of problem they would like to discuss with the panel.

Second Workshop On Self-Organization In Representations For Evolutionary Algorithms: Building complexity from simplicity
Ivan I. Garibay, Sanjeev Kumar, Ozlem Garibay, Hal Stringer
http://ivan.research.ucf.edu/SOEA-2005.htm

Summary:
The success of Evolutionary Algorithms (EAs) on a wide range of otherwise intractable problems has promoted its use. As EAs are applied to increasingly difficult problems that require increasingly complex solutions, they face a number of problems: premature convergence to suboptimal solutions, stagnation of search in large search spaces, negative epistatic effects, disruption of large building blocks, and scalability, among others. Nature evolves instructions in the form of genes that are used to specify the construction of organisms using a highly non-linear process: development.
Self-organization is fundamental to the developmental process at all levels: molecular, genetic, and cellular. With new reports of the number of genes in the human genome being revised downwards, the role< of self-organization in complex webs of gene regulation is all the more salient. Given these new findings, perhaps the self-organization of genotypic instructions that transform genotype to phenotype is a key missing ingredient necessary for unleashing the evolution of complex and scalable solutions with emergent phenomena such as: scale-free-ness, adaptability, innovation, evolvability, and robustness.This workshop will focus on domain-independent methods for representing complex solutions with self-organizable building blocks, and on developmental principles for specifying the construction of complex systems. The workshop welcomes submissions from biologists on relevant biology that may help shed more light on self-organizing principles for evolutionary computation.

Topics of interest include:

* Models of complexity building using self-organization
* Emergent behavior in representations
* Methods of design and evaluation of self-organizable representational
building blocks
* Scalability of self-organizational processes to high complexities
* Self-organization theoretical approaches: complexity, chaos, synergetics,
self-organized criticality, non-equilibrium thermodynamics, etc.
* Self-organized development
* Genotype-phenotype mappings for self-organization and single & multicellular
development
* Pattern formation, morphogenesis, cellular differentiation, and growth
* Models of genetic regulatory networks, modularity, segmentation, and
compartmentalization
* Scalability & Evolvability of developmental processes
* Robustness, self-repair and regeneration in developmental processes
* Real world applications of developmental principles

Theory of Representations
Marc Toussaint, Alden H. Wright, Edwin D. de Jong
http://homepages.inf.ed.ac.uk/mtoussai/gecco05/

The choice of representation crucially determines the performance of a heuristic search process. We believe that there have been very interesting new ideas and approaches on the subject of learning representations recently. However, a unifying point of view is currently missing and the different approaches are widely scattered in the literature with too little cross-fertilization. In this workshop we would like to gather such work and, in a discussion between the contributors from the various lines of research, fuse the various approaches and formalisms into a common framework. This framework might clarify what the scope of a theory of representations should be, what existing algorithms may be considered as cases of representation adaptation, and how the existing literature on the topic can be integrated in a broader picture -- thereby also seeking contact with related areas in Computer Science and Machine Learning.

The workshop will be a half-day event, during which approximately eight undergraduate students will present their work to a panel of GECCO participants interested in undergraduate education. The panel will also include participating students' faculty mentors. Students should plan on 15-minute presentations, followed by five minutes of questions. The panel will provide feedback to the presenting students regarding their work and their presentation.

Students invited to the workshop will also participate in the conference poster session. Students will display posters summarizing their work; this will allow the larger GECCO community to see what's being done by undergraduates in evolutionary computation. The poster session will also be a great opportunity for some networking!

Goals of the Undergraduate Student Workshop include:

(*) To provide a forum allowing undergraduate students to put a " capstone" on their undergraduate research activities, by presenting their work at an international conference

(*) To encourage teaching faculty to consider undergraduate research opportunities for their students in the EC field

(*) To help prepare undergraduate students for graduate work in EC areas

(*) To encourage sharing and networking amongst teaching faculty with students participating in undergraduate research projects in EC

(*) To provide networking opportunities for graduate school faculty and undergraduate students interested in pursuing advanced degrees, and

(*) To encourage more emphasis on education at the GECCO conference