Developing Complex Systems Using Evolved Pattern Generators
Active from 2004 - 2006
Self-organization of connection patterns within brain areas of animals begins prenatally, and has been shown to depend on internally generated patterns of neural activity. The neural structures continue to develop postnatally through externally driven patterns, when the sensory systems are exposed to stimuli from the environment. The internally generated patterns have been proposed to give the neural system an appropriate bias so that it can learn reliably from complex environmental stimuli. This project evaluates the hypothesis that complex artificial learning systems can benefit from a similar approach, consisting of initial training with patterns from an evolved pattern generator, followed by training with the actual training set. Preliminary experiments to test this hypothesis involved training competitive learning networks for recognizing handwritten digits. The results demonstrate how the approach can improve learning performance by discovering the appropriate initial weight biases, thereby compensating for weaknesses of the learning algorithm.
Vinod Valsalam Ph.D. Alumni vkv [at] alumni utexas net
James A. Bednar Postdoctoral Alumni jbednar [at] inf ed ac uk
Risto Miikkulainen Faculty risto [at] cs utexas edu
Developing Complex Systems Using Evolved Pattern Generators Vinod K. Valsalam, James A. Bednar and Risto Miikkulainen IEEE Transactions on Evolutionary Computation:181-198, 2007. 2007

Establishing an Appropriate Learning Bias Through Development Vinod K. Valsalam, James A. Bednar, and Risto Miikkulainen In Proceedings of the Fifth International Conference on Development and Learning (ICDL-2006),... 2006

Constructing Good Learners Using Evolved Pattern Generators Vinod K. Valsalam, James A. Bednar, and Risto Miikkulainen In H.-G. Beyer and others, editors, Proceedings of the Genetic and Evolutionary Computation Confe... 2005