Connectionists: New Experiment for Neural-Inspired Evolution of Artwork

[Jimmy Secretan]

I would like to post the following announcement to the list about a 
neural-inspired interactive evolution experiment we would like to 
announce the Connectionist community:

[BEGIN ANNOUNCEMENT]

Dear Colleagues,

The Evolutionary Complexity Group (E-plex) (http://eplex.cs.ucf.edu) at 
the University of Central Florida is pleased to announce a new 
interactive website called Picbreeder (http://www.picbreeder.org), which 
is a large-scale on-line experiment in Collaborative Interactive 
Evolution (CIE) of artwork through neural networks. We invite the 
Connectionists community to visit and participate.

Images on Picbreeder are rendered by kinds of neural networks called
Compositional Pattern Producing Networks (CPPNs), which are in turn 
evolved by the NeuroEvolution of Augmenting Topologies (NEAT) algorithm. 
  Users can select and breed the CPPNs that render the most appealing 
images, and publish them to the community.

Most importantly, users can continue evolving others' CPPNs, encouraging 
a continual increase in image complexity. As a result, users have been 
able to evolve genuinely recognizable objects such as faces, cars, and 
animals, all without any a priori knowledge in the system, a novel 
achievement for neural-inspired methods.

Please see our recent short publication on Picbreeder for more information:

http://eplex.cs.ucf.edu/index.php?option=com_content&task=view&id=14&Itemid=28#secretan.leonardotrans07 

Picbreeder: Collaborative Interactive Evolution of Images
Jimmy Secretan, Nicholas Beato, David B. D'Ambrosio, Adelein Rodriguez, 
Adam Campbell and Kenneth O. Stanley
In: Leonardo (Transactions Section) Vol. 41, No. 1 (2007)

[END ANNOUNCEMENT]

Thank you.

Sincerely,

	Jimmy Secretan