Volltext-Downloads (blau) und Frontdoor-Views (grau)

Prediction of neural network performance by phenotypic modeling

  • Surrogate models are used to reduce the burden of expensive-to-evaluate objective functions in optimization. By creating models which map genomes to objective values, these models can estimate the performance of unknown inputs, and so be used in place of expensive objective functions. Evolutionary techniques such as genetic programming or neuroevolution commonly alter the structure of the genome itself. A lack of consistency in the genotype is a fatal blow to data-driven modeling techniques: interpolation between points is impossible without a common input space. However, while the dimensionality of genotypes may differ across individuals, in many domains, such as controllers or classifiers, the dimensionality of the input and output remains constant. In this work we leverage this insight to embed differing neural networks into the same input space. To judge the difference between the behavior of two neural networks, we give them both the same input sequence, and examine the difference in output. This difference, the phenotypic distance, can then be used to situate these networks into a common input space, allowing us to produce surrogate models which can predict the performance of neural networks regardless of topology. In a robotic navigation task, we show that models trained using this phenotypic embedding perform as well or better as those trained on the weight values of a fixed topology neural network. We establish such phenotypic surrogate models as a promising and flexible approach which enables surrogate modeling even for representations that undergo structural changes.

Export metadata

Additional Services

Search Google Scholar Check availability


Show usage statistics
Document Type:Conference Object
Author:Alexander Hagg, Martin Zaefferer, Jörg Stork, Adam Gaier
Parent Title (English):GECCO '19: Genetic and Evolutionary Computation Conference, Prague, Czech Republic, July 13-17, 2019
First Page:1576
Last Page:1582
ArXiv Id:http://arxiv.org/abs/1907.07075
Publisher:Association for Computing Machinery
Place of publication:New York, NY, United States
Date of first publication:2019/07/13
Copyright:© 2019 Copyright held by the owner/author(s). Publication rights licensed to the Association for Computing Machinery. Abstracting with credit is permitted.
Departments, institutes and facilities:Fachbereich Informatik
Institut für Technik, Ressourcenschonung und Energieeffizienz (TREE)
Projects:AErOMAt - Automatisiertes Entwickeln aerodynamischer Strukturen und Fahrzeuge mithilfe evolutionärer Optimierung und Surrogatmodellierung (DE/BMBF/03FH012PX5,13FH012PX5)
Dewey Decimal Classification (DDC):0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 004 Datenverarbeitung; Informatik
Entry in this database:2019/07/13