State prediction: a constructive method to program recurrent neural networks

TitleState prediction: a constructive method to program recurrent neural networks
Publication TypeConference Paper
Year of Publication2011
AuthorsReinhart F, Steil JJ
Conference NameArtificial Neural Networks and Machine Learning – ICANN 2011 : 21st International Conference on Artificial Neural Networks, Espoo, Finland, June 14-17, 2011, Proceedings, Part I
PublisherSpringer
Abstract

We introduce a novel technique to program desired state sequences into recurrent neural networks in one shot. The basic methodology and its scalability to large and input-driven networks is demonstrated by shaping attractor landscapes, transient dynamics and programming limit cycles. The approach unifies programming of transient and attractor dynamics in a generic framework.

DOI10.1007/978-3-642-21735-7_20