TY - GEN

T1 - Evolving neural networks in compressed weight space

AU - Koutník, Jan

AU - Gomez, Faustino

AU - Schmidhuber, Jürgen

N1 - Generated from Scopus record by KAUST IRTS on 2022-09-14

PY - 2010/8/27

Y1 - 2010/8/27

N2 - We propose a new indirect encoding scheme for neural networks in which the weight matrices are represented in the frequency domain by sets of Fourier coefficients. This scheme exploits spatial regularities in the matrix to reduce the dimensionality of the representation by ignoring high-frequency coefficients, as is done in lossy image compression. We compare the efficiency of searching in this "compressed" network space to searching in the space of directly encoded networks, using the CoSyNE neuroevolution algorithm on three benchmark problems: pole-balancing, ball throwing and octopusarm control. The results show that this encoding can dramatically reduce the search space dimensionality such that solutions can be found in significantly fewer evaluations. Copyright 2010 ACM.

AB - We propose a new indirect encoding scheme for neural networks in which the weight matrices are represented in the frequency domain by sets of Fourier coefficients. This scheme exploits spatial regularities in the matrix to reduce the dimensionality of the representation by ignoring high-frequency coefficients, as is done in lossy image compression. We compare the efficiency of searching in this "compressed" network space to searching in the space of directly encoded networks, using the CoSyNE neuroevolution algorithm on three benchmark problems: pole-balancing, ball throwing and octopusarm control. The results show that this encoding can dramatically reduce the search space dimensionality such that solutions can be found in significantly fewer evaluations. Copyright 2010 ACM.

UR - http://portal.acm.org/citation.cfm?doid=1830483.1830596

UR - http://www.scopus.com/inward/record.url?scp=77955872473&partnerID=8YFLogxK

U2 - 10.1145/1830483.1830596

DO - 10.1145/1830483.1830596

M3 - Conference contribution

SN - 9781450300728

SP - 619

EP - 625

BT - Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10

ER -