Self-organizing maps with dynamic learning for signal reconstruction

Jeongho Cho; António R.C. Paiva; Sung Phil Kim; Justin C. Sanchez; José C. Príncipe

doi:10.1016/j.neunet.2006.12.002

Self-organizing maps with dynamic learning for signal reconstruction

Jeongho Cho, António R.C. Paiva, Sung Phil Kim, Justin C. Sanchez, José C. Príncipe

Biomedical Engineering

Research output: Contribution to journal › Article › peer-review

11 Scopus citations

Abstract

Wireless Brain Machine Interface (BMI) communication protocols are faced with the challenge of transmitting the activity of hundreds of neurons which requires large bandwidth. Previously a data compression scheme for neural activity was introduced based on Self Organizing Maps (SOM). In this paper we propose a dynamic learning rule for improved training of the SOM on signals with sparse events which allows for more representative prototype vectors to be found, and consequently better signal reconstruction. This work was developed with BMI applications in mind and therefore our examples are geared towards this type of signals. The simulation results show that the proposed strategy outperforms conventional vector quantization methods for spike reconstruction.

Original language	English (US)
Pages (from-to)	274-284
Number of pages	11
Journal	Neural Networks
Volume	20
Issue number	2
DOIs	https://doi.org/10.1016/j.neunet.2006.12.002
State	Published - Mar 2007

Keywords

Brain-machine interface
Self-organizing map
Spike reconstruction
Vector quantization

ASJC Scopus subject areas

Artificial Intelligence
Neuroscience(all)

Access to Document

10.1016/j.neunet.2006.12.002

Fingerprint Dive into the research topics of 'Self-organizing maps with dynamic learning for signal reconstruction'. Together they form a unique fingerprint.

Cite this

@article{de73559cecbc40c0b8cdd7c9b909b7b6,

title = "Self-organizing maps with dynamic learning for signal reconstruction",

abstract = "Wireless Brain Machine Interface (BMI) communication protocols are faced with the challenge of transmitting the activity of hundreds of neurons which requires large bandwidth. Previously a data compression scheme for neural activity was introduced based on Self Organizing Maps (SOM). In this paper we propose a dynamic learning rule for improved training of the SOM on signals with sparse events which allows for more representative prototype vectors to be found, and consequently better signal reconstruction. This work was developed with BMI applications in mind and therefore our examples are geared towards this type of signals. The simulation results show that the proposed strategy outperforms conventional vector quantization methods for spike reconstruction.",

keywords = "Brain-machine interface, Self-organizing map, Spike reconstruction, Vector quantization",

author = "Jeongho Cho and Paiva, {Ant{\'o}nio R.C.} and Kim, {Sung Phil} and Sanchez, {Justin C.} and Pr{\'i}ncipe, {Jos{\'e} C.}",

note = "Funding Information: This work was partially supported by DARPA under grant ONR-450595112. Ant{\'o}nio R. C. Paiva was supported by Funda{\c c}{\~a}o para a Ci{\^e}ncia e a Tecnologia under grant SFRH/BD/18217/2004. Copyright: Copyright 2008 Elsevier B.V., All rights reserved.",

year = "2007",

month = mar,

doi = "10.1016/j.neunet.2006.12.002",

language = "English (US)",

volume = "20",

pages = "274--284",

journal = "Neural Networks",

issn = "0893-6080",

publisher = "Elsevier Limited",

number = "2",

}

TY - JOUR

T1 - Self-organizing maps with dynamic learning for signal reconstruction

AU - Cho, Jeongho

AU - Paiva, António R.C.

AU - Kim, Sung Phil

AU - Sanchez, Justin C.

AU - Príncipe, José C.

N1 - Funding Information: This work was partially supported by DARPA under grant ONR-450595112. António R. C. Paiva was supported by Fundação para a Ciência e a Tecnologia under grant SFRH/BD/18217/2004. Copyright: Copyright 2008 Elsevier B.V., All rights reserved.

PY - 2007/3

Y1 - 2007/3

N2 - Wireless Brain Machine Interface (BMI) communication protocols are faced with the challenge of transmitting the activity of hundreds of neurons which requires large bandwidth. Previously a data compression scheme for neural activity was introduced based on Self Organizing Maps (SOM). In this paper we propose a dynamic learning rule for improved training of the SOM on signals with sparse events which allows for more representative prototype vectors to be found, and consequently better signal reconstruction. This work was developed with BMI applications in mind and therefore our examples are geared towards this type of signals. The simulation results show that the proposed strategy outperforms conventional vector quantization methods for spike reconstruction.

AB - Wireless Brain Machine Interface (BMI) communication protocols are faced with the challenge of transmitting the activity of hundreds of neurons which requires large bandwidth. Previously a data compression scheme for neural activity was introduced based on Self Organizing Maps (SOM). In this paper we propose a dynamic learning rule for improved training of the SOM on signals with sparse events which allows for more representative prototype vectors to be found, and consequently better signal reconstruction. This work was developed with BMI applications in mind and therefore our examples are geared towards this type of signals. The simulation results show that the proposed strategy outperforms conventional vector quantization methods for spike reconstruction.

KW - Brain-machine interface

KW - Self-organizing map

KW - Spike reconstruction

KW - Vector quantization

UR - http://www.scopus.com/inward/record.url?scp=33847111116&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33847111116&partnerID=8YFLogxK

U2 - 10.1016/j.neunet.2006.12.002

DO - 10.1016/j.neunet.2006.12.002

M3 - Article

C2 - 17234384

AN - SCOPUS:33847111116

VL - 20

SP - 274

EP - 284

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

IS - 2

ER -

Self-organizing maps with dynamic learning for signal reconstruction

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Cite this