Mitigating Architectural Mismatch During the Evolutionary Synthesis of Deep Neural Networks

  • Audrey G. Chung
  • Paul Feiguth
  • Alexander Wong

Abstract

Evolutionary deep intelligence has recently shown great promise
for producing small, powerful deep neural network models via the
organic synthesis of increasingly efficient architectures over suc-
cessive generations. Existing evolutionary synthesis processes,
however, have allowed the mating of parent networks independent
of architectural alignment, resulting in a mismatch of network struc-
tures. We present a preliminary study into the effects of architec-
tural alignment during evolutionary synthesis using a gene tagging
system. Surprisingly, the network architectures synthesized using
the gene tagging approach resulted in slower decreases in perfor-
mance accuracy and storage size; however, the resultant networks
were comparable in size and performance accuracy to the non-
gene tagging networks. Furthermore, we speculate that there is a
noticeable decrease in network variability for networks synthesized
with gene tagging, indicating that enforcing a like-with-like mating
policy potentially restricts the exploration of the search space of
possible network architectures.

Published
2018-12-24
How to Cite
Chung, A., Feiguth, P., & Wong, A. (2018). Mitigating Architectural Mismatch During the Evolutionary Synthesis of Deep Neural Networks. Journal of Computational Vision and Imaging Systems, 4(1), 3. Retrieved from https://openjournals.uwaterloo.ca/index.php/vsl/article/view/323