Utilizing Symmetry in Evolutionary Design
Active from 2007 - 2010
Can symmetry be utilized as a design principle to constrain evolutionary search, making it more effective? This dissertation aims to show that this is indeed the case, in two ways. First, an approach called ENSO is developed to evolve modular neural network controllers for simulated multilegged robots. Inspired by how symmetric organisms have evolved in nature, ENSO utilizes group theory to break symmetry systematically, constraining evolution to explore promising regions of the search space. As a result, it evolves effective controllers even when the appropriate symmetry constraints are difficult to design by hand. The controllers perform equally well when transferred from simulation to a physical robot. Second, the same principle is used to evolve minimal-size sorting networks. In this different domain, a different instantiation of the same principle is effective: building the desired symmetry step-by-step. This approach is more scalable than previous methods and finds smaller networks, thereby demonstrating that the principle is general. Thus, evolutionary search that utilizes symmetry constraints is shown to be effective in a range of challenging applications.
Vinod Valsalam Ph.D. Alumni vkv [at] alumni utexas net
Using Symmetry and Evolutionary Search to Minimize Sorting Networks Vinod K. Valsalam and Risto Miikkulainen Journal of Machine Learning Research, 14(Feb):303--331, 2013. 2013

Constructing Controllers for Physical Multilegged Robots using the ENSO Neuroevolution Approach Vinod K. Valsalam, Jonathan Hiller, Robert MacCurdy, Hod Lipson and Risto Miikkulainen Evolutionary Intelligence, 5(1):1--12, 2012. 2012

Evolving Symmetry for Modular System Design Vinod K. Valsalam and Risto Miikkulainen IEEE Transactions on Evolutionary Computation, 15(3):368--386, 2011. 2011

Utilizing Symmetry and Evolutionary Search to Minimize Sorting Networks Vinod K. Valsalam and Risto Miikkulainen Technical Report AITR-11-09, Department of Computer Sciences, The University of Texas at Austin, Aus... 2011

Utilizing Symmetry in Evolutionary Design Vinod Valsalam PhD Thesis, Department of Computer Sciences, The University of Texas at Austin, Austin, TX, 2010. Te... 2010

Evolving Symmetric and Modular Neural Network Controllers for Multilegged Robots Vinod K. Valsalam and Risto Miikkulainen In xploring New Horizons in Evolutionary Design of Robots: Workshop at the 2009 IEEE/RSJ Internat... 2009

Evolving Symmetric and Modular Neural Networks for Distributed Control Vinod K. Valsalam and Risto Miikkulainen In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO), 731--738, 200... 2009

Modular Neuroevolution for Multilegged Locomotion Vinod K. Valsalam and Risto Miikkulainen In Proceedings of the Genetic and Evolutionary Computation Conference GECCO 2008, 265-272, Ne... 2008

ENSO This package contains software implementing the ENSO approach for evolving symmetric modular neural networks. It also in... 2010

Sorting Networks This package contains software utilizing an approach based on symmetry and evolution to minimize the number of comparato... 2010