Assessing Architectural Similarity in Populations of Deep Neural Networks

Evolutionary deep intelligence has recently shown great promise for producing small, powerful deep neural network models via the synthesis of increasingly efficient architectures over successive generations. Despite recent research showing the efficacy of multi-parent evolutionary synthesis, little...

Full description

Saved in:
Bibliographic Details
Main Authors: Chung, Audrey G., Fieguth, Paul, Wong, Alexander
Format: Conference Proceeding
Language:English
Subjects:
Network architecture
Neural networks
Pose estimation
Sociology
Speech recognition
Statistics
Tagging
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Evolutionary deep intelligence has recently shown great promise for producing small, powerful deep neural network models via the synthesis of increasingly efficient architectures over successive generations. Despite recent research showing the efficacy of multi-parent evolutionary synthesis, little has been done to directly assess architectural similarity between networks during the synthesis process for improved parent network selection. In this work, we present a preliminary study into quantifying architectural similarity via the percentage overlap of architectural clusters. Results show that networks synthesized using architectural alignment (via gene tagging) maintain higher architectural similarities within each generation, potentially restricting the search space of highly efficient network architectures.
ISSN:2160-7516
DOI:10.1109/CVPRW.2019.00070