MULTIOBJECTIVE COEVOLUTION OF DEEP NEURAL NETWORK ARCHITECTURES

There is described an automated machine learning (AutoML) implementation system, comprising: an algorithm layer for evolving deep neural network (DNN) hyperparameters and architectures; a system layer for parallel training of multiple evolved DNNs received from the algorithm layer, detennination of...

Full description

Saved in:
Bibliographic Details
Main Authors MIIKKULAINEN, RISTO, LIANG, JASON ZHI, MEYERSON, ELLIOTT
Format Patent
LanguageEnglish
French
Published 07.05.2020
Online AccessGet full text

Cover

More Information
Summary:There is described an automated machine learning (AutoML) implementation system, comprising: an algorithm layer for evolving deep neural network (DNN) hyperparameters and architectures; a system layer for parallel training of multiple evolved DNNs received from the algorithm layer, detennination of one or more fitness values for each of the received DNNs, and providing the one or more fitness values back to the algorithm layer for use in additional generations of evolving DNNs; and a program layer for infonning the algorithm layer and system layer of one or more desired optimization characteristics of the evolved DNNs.
Bibliography:Application Number: CA20193215345