Header

UZH-Logo

Maintenance Infos

A morphological learning: Increased memory capacity of neuromorphic systems with binary synapses exploiting AER based reconfiguration


Hussain, S; Gopalakrishnan, R; Basu, A; Liu, S-C (2013). A morphological learning: Increased memory capacity of neuromorphic systems with binary synapses exploiting AER based reconfiguration. In: The International Joint Conference on Neural Networks (IJCNN) 2013, Dallas, TX, USA, 4 August 2013 - 9 August 2013, 1-7.

Abstract

Spiking neurons with lumped nonlinearity representing active dendrites can perform a larger number of input-output mappings than is possible by a neuron with linear synaptic summation of its currents. This is possible due to the additional degree of freedom in such cells-its `morphology' reflected in the number of dendrites and the choice of which inputs form synapses on the same dendrite. We present a hardware friendly algorithm for learning such optimal morphologies utilizing correlations between inputs and dendritic branch activations. We demonstrate the increased memory capacity of neurons with nonlinear dendrites and binary synapses over typically used linearly summing cells with high resolution weights. We have shown that a neuron model with a fixed number of binary weights performs much worse on a pattern classification task when it uses traditional linear dendrites than when it utilizes nonlinear dendrites (19% compared to 9% errors for 1000 patterns). This method allows to trade-off weight resolution, a problem in most current neuromorphic systems, with configurability that is the strength of address event representation (AER) based systems which can store configuration details in an off-chip memory. On a fundamental level, it points to the need of having a higher ratio of nonlinear to linear operations in spiking neural networks than is typically used.

Abstract

Spiking neurons with lumped nonlinearity representing active dendrites can perform a larger number of input-output mappings than is possible by a neuron with linear synaptic summation of its currents. This is possible due to the additional degree of freedom in such cells-its `morphology' reflected in the number of dendrites and the choice of which inputs form synapses on the same dendrite. We present a hardware friendly algorithm for learning such optimal morphologies utilizing correlations between inputs and dendritic branch activations. We demonstrate the increased memory capacity of neurons with nonlinear dendrites and binary synapses over typically used linearly summing cells with high resolution weights. We have shown that a neuron model with a fixed number of binary weights performs much worse on a pattern classification task when it uses traditional linear dendrites than when it utilizes nonlinear dendrites (19% compared to 9% errors for 1000 patterns). This method allows to trade-off weight resolution, a problem in most current neuromorphic systems, with configurability that is the strength of address event representation (AER) based systems which can store configuration details in an off-chip memory. On a fundamental level, it points to the need of having a higher ratio of nonlinear to linear operations in spiking neural networks than is typically used.

Statistics

Citations

Altmetrics

Downloads

0 downloads since deposited on 12 Feb 2014
0 downloads since 12 months

Additional indexing

Item Type:Conference or Workshop Item (Speech), refereed, original work
Communities & Collections:07 Faculty of Science > Institute of Neuroinformatics
Dewey Decimal Classification:570 Life sciences; biology
Language:English
Event End Date:9 August 2013
Deposited On:12 Feb 2014 16:58
Last Modified:08 Dec 2017 03:24
Publisher:Proceedings of the International Joint Conference on Neural Networks 2013
Series Name:2013 International Joint Conference on Neural Networks
Publisher DOI:https://doi.org/10.1109/IJCNN.2013.6706928

Download