Dr. Jeff CluneThe KurzweilAI.net article Artificial Life Forms Evolve Basic Intelligence said
Michigan State University (MSU) researchers have developed “digital organisms” called Avidians that were made to evolve memory, and could eventually be used to generate intelligent artificial life and evolve into symmetrical, organized artificial brains that share structural properties with real brains.
MSU researcher Jeff Clune works with a system called HyperNEAT, which uses principles of developmental biology to grow a large number of digital neurons from a small number of instructions. He translated the artificial neurons into code that could control a Roomba robot.
You can build complex brains from a relatively small number of computerized instructions, or “genes”, he says. Their brains have millions of connections, yet still perform a task well, and that number could be pushed higher yet. “This is a sea change for the field. Being able to evolve functional brains at this scale allows us to begin pushing the capabilities of artificial neural networks up, and opens up a path to evolving artificial brains that rival their natural counterparts.”
Jeff Clune, Ph.D. is Postdoctoral Fellow at Cornell University,
working with professor Hod Lipson.
He is Co-chair, Generative and Developmental Systems Track, Genetic and
Evolutionary Computation Conference, 2011 and winner of the National
Science Foundation’s Postdoctoral Research Fellowship in Biology award.
His publications include: Natural Selection Fails to Optimize Mutation Rates for Long-Term Adaptation on Rugged Fitness Landscapes, Investigating Whether HyperNEAT Produces Modular Neural Networks, HybrID: A hybridization of indirect and direct encodings for evolutionary computation, The Evolution of Division of Labor, The Sensitivity of HyperNEAT to Different Geometric Representations of a Problem, and Problem Decomposition Using Indirect Reciprocity in Evolved Populations. Read the full list of his publications!
Jeff was one of three developers that wrote a software package called Avida-ED, which is used in university biology classes to teach evolution. With it, students are able to conduct research in experimental evolution in real time. It is a great way to allow students to watch evolution happen and get hands on experience in seeing how evolution behaves in different situations. It also helps students learn about the nature of scientific inquiry itself.
Avida-ED was discussed in Science magazine and has been used in classrooms in many universities throughout the world. The NSF grant overseer for the Avida-ED project described it as “one of the most successful science education materials projects with which I am acquainted. The product is excellent, dissemination is already successful, and the assessment plan is outstanding.”
Jeff earned his Ph.D. in Computer Science at Michigan State University with the dissertation Evolving Artificial Neural Networks with Generative Encodings Inspired by Developmental Biology in 2010. He earned his Masters in Philosophy at Michigan State University in 2005 and his B.A. in Philosophy (with honors) at University of Michigan in 1999.
Listen to his Biota Live Podcast.