Evolving neural networks to follow trajectories of arbitrary complexity

Inden, B. ORCID: 0000-0001-6048-6856 and Jost, J., 2019. Evolving neural networks to follow trajectories of arbitrary complexity. Neural Networks, 116, pp. 224-236. ISSN 0893-6080

13976_Inden.pdf - Post-print

Download (4MB) | Preview


Many experiments have been performed that use evolutionary algorithms for learning the topology and connection weights of a neural network that controls a robot or virtual agent. These experiments are not only performed to better understand basic biological principles, but also with the hope that with further progress of the methods, they will become competitive for automatically creating robot behaviors of interest. However, current methods are limited with respect to the (Kolmogorov) complexity of evolved behavior. Using the evolution of robot trajectories as an example, we show that by adding four features, namely (1) freezing of previously evolved structure, (2) temporal scaffolding, (3) a homogeneous transfer function for output nodes, and (4) mutations that create new pathways to outputs, to standard methods for the evolution of neural networks, we can achieve an approximately linear growth of the complexity of behavior over thousands of generations. Overall, evolved complexity is up to two orders of magnitude over that achieved by standard methods in the experiments reported here, with the major limiting factor for further growth being the available run time. Thus, the set of methods proposed here promises to be a useful addition to various current neuroevolution methods.

Item Type: Journal article
Publication Title: Neural Networks
Creators: Inden, B. and Jost, J.
Publisher: Pergamon Press
Date: August 2019
Volume: 116
ISSN: 0893-6080
S089360801930111XPublisher Item Identifier
Record created by: Jonathan Gallacher
Date Added: 29 May 2019 08:57
Last Modified: 08 May 2020 03:00
URI: https://irep.ntu.ac.uk/id/eprint/36667

Actions (login required)

Edit View Edit View


Views per month over past year


Downloads per month over past year