Connectionists: Announcing ES-HyperNEAT: An Enhanced Hypercube-Based Encoding for Evolving the Placement, Density, and Connectivity of Neurons

[Sebastian Risi]

Dear Connectionists,

The recently introduced Hypercube-based NeuroEvolution of Augmenting
Topologies (HyperNEAT) is a step beyond traditional neural network
evolution (i.e. neuroevolution) algorithms towards evolving more
brain-like structures through evolutionary algorithms. In particular,
neural networks evolved by HyperNEAT feature topography in addition to
topology. That is, neurons exist at spatial locations just as they do
in real brains, which means that connectivity patterns evolve that can
be analyzed for emergent topographic map-like characteristics. In
addition, the ability to encode and evolve large connectivity patterns
with regularities means that HyperNEAT can evolve larger networks than
past approaches, with up to millions of connections. Yet the positions
and number of the neurons connected through this approach must be
decided a priori by the user and, unlike in living brains, cannot
change during evolution.

We are pleased to announce a new paper that introduces
Evolvable-substrate HyperNEAT (ES-HyperNEAT), which addresses this
limitation by automatically deducing the density and positions of
neurons from implicit information in the pattern of weights encoded by
HyperNEAT, thereby avoiding the need to evolve explicit placement.
This approach not only can evolve the location of every neuron in the
network (while still preserving the advances introduced by the
original HyperNEAT), but also can represent regions of varying
density, which means resolution can increase holistically over
evolution. In this paper, we show that ES-HyperNEAT can significantly
expand the scope of neural structures that evolution can discover.

Cite: An Enhanced Hypercube-Based Encoding for Evolving the Placement,
Density, and Connectivity of Neurons. Sebastian Risi and Kenneth O.
Stanley. Artificial Life journal, Vol. 18, No. 4, Pages 331-363. MIT
Press, 2012

http://www.mitpressjournals.org/doi/pdf/10.1162/ARTL_a_00071

Manuscript: http://eplex.cs.ucf.edu/publications/2012/risi-alife12

In the past four years, a significant body of research from a growing
HyperNEAT community has emerged. Many of these publications, source
code, and a short online introduction to the technique are available
at the HyperNEAT Users Page:

http://eplex.cs.ucf.edu/hyperNEATpage/HyperNEAT.html

Additional ES-HyperNEAT specific information is available here:

http://eplex.cs.ucf.edu/ESHyperNEAT/

Best,

Sebastian Risi

-- 
Dr. Sebastian Risi
Postdoctoral Fellow
Creative Machines Laboratory
Cornell University
Email: sebastian.risi at cornell.edu  Tel: (407) 929-5113
Web: http://www.cs.ucf.edu/~risi/