Start date should be earlier than end date. OK Please enter valid date format.

Original Research ARTICLE

This article is part of a Research Topic

SORN: a self-organizing recurrent neural network

Andreea Lazar¹*, Gordon Pipa^1,2 and Jochen Triesch¹

Frankfurt Institute of Advanced Studies, Johann Wolfgang Goethe University, Frankfurt am Main, Germany

Department of Neurophysiology, Max Planck Institute for Brain Research, Frankfurt am Main, Germany

Understanding the dynamics of recurrent neural networks is crucial for explaining how the brain processes information. In the neocortex, a range of different plasticity mechanisms are shaping recurrent networks into effective information processing circuits that learn appropriate representations for time-varying sensory stimuli. However, it has been difficult to mimic these abilities in artificial neural network models. Here we introduce SORN, a self-organizing recurrent network. It combines three distinct forms of local plasticity to learn spatio-temporal patterns in its input while maintaining its dynamics in a healthy regime suitable for learning. The SORN learns to encode information in the form of trajectories through its high-dimensional state space reminiscent of recent biological findings on cortical coding. All three forms of plasticity are shown to be essential for the network’s success.

Keywords:

synaptic plasticity, intrinsic plasticity, recurrent neural networks, reservoir computing, time series prediction

Citation:

Lazar A, Pipa G and Triesch J (2009). SORN: a self-organizing recurrent neural network. Front. Comput. Neurosci. 3:23. doi: 10.3389/neuro.10.023.2009

Received:

24 June 2009;

Paper pending published:

04 August 2009;

Accepted:

05 October 2009;

Published online:

30 October 2009.

Edited by:

Hava T. Siegelmann, University of Massachusetts Amherst, USA

Reviewed by:

Phil Goodman, University of Nevada School of Medicine, USA
Robert Kozma, University of Memphis, USA

© 2009 Lazar, Pipa and Triesch.This is an open-access article subject to an exclusive license agreement between the authors and the Frontiers Research Foundation, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are credited.

*Correspondence:

Andreea Lazar, Frankfurt Institute for Advanced Studies, Ruth-Moufang-Str.1, 60438 Frankfurt am Main, Germany.e-mail: lazar@fias.uni-frankfurt.de