Visualising error surfaces for adaptive filters and other purposes

MH Fisher; DP Mandic; JA Bangham; RW Harvey

doi:10.1109/ICASSP.2000.860161

Visualising error surfaces for adaptive filters and other purposes

MH Fisher, DP Mandic, JA Bangham, RW Harvey

Research output: Contribution to conference › Other

6 Citations (Scopus)

Abstract

Modern neural and adaptive systems often have complicated error performance surfaces with many local extrema. Visualising and understanding these surfaces is critical to effective tuning of these systems but almost all visualisation methods are confined to two dimensions. Here we show how to use a morphological scale-space transform to convert these multi-dimensional complex error surfaces into two-dimensional trees where the leaf nodes are local minima and other nodes represent decision points such as saddle points and points of inflection.

Original language	English
Pages	3522-3525
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2000.860161
Publication status	Published - 2000
Event	IEEE International Conference Acoustics Speech and Signal Processing, ICASSP2000 - Istanbul, Turkey Duration: 5 Jun 2000 → 9 Jun 2000

Conference

Conference	IEEE International Conference Acoustics Speech and Signal Processing, ICASSP2000
Country/Territory	Turkey
City	Istanbul
Period	5/06/00 → 9/06/00

Access to Document

10.1109/ICASSP.2000.860161

Cite this

@conference{45495717015944e78a6396de4e092003,

title = "Visualising error surfaces for adaptive filters and other purposes",

abstract = "Modern neural and adaptive systems often have complicated error performance surfaces with many local extrema. Visualising and understanding these surfaces is critical to effective tuning of these systems but almost all visualisation methods are confined to two dimensions. Here we show how to use a morphological scale-space transform to convert these multi-dimensional complex error surfaces into two-dimensional trees where the leaf nodes are local minima and other nodes represent decision points such as saddle points and points of inflection.",

author = "MH Fisher and DP Mandic and JA Bangham and RW Harvey",

year = "2000",

doi = "10.1109/ICASSP.2000.860161",

language = "English",

pages = "3522--3525",

note = "IEEE International Conference Acoustics Speech and Signal Processing, ICASSP2000 ; Conference date: 05-06-2000 Through 09-06-2000",

}

TY - CONF

T1 - Visualising error surfaces for adaptive filters and other purposes

AU - Fisher, MH

AU - Mandic, DP

AU - Bangham, JA

AU - Harvey, RW

PY - 2000

Y1 - 2000

N2 - Modern neural and adaptive systems often have complicated error performance surfaces with many local extrema. Visualising and understanding these surfaces is critical to effective tuning of these systems but almost all visualisation methods are confined to two dimensions. Here we show how to use a morphological scale-space transform to convert these multi-dimensional complex error surfaces into two-dimensional trees where the leaf nodes are local minima and other nodes represent decision points such as saddle points and points of inflection.

AB - Modern neural and adaptive systems often have complicated error performance surfaces with many local extrema. Visualising and understanding these surfaces is critical to effective tuning of these systems but almost all visualisation methods are confined to two dimensions. Here we show how to use a morphological scale-space transform to convert these multi-dimensional complex error surfaces into two-dimensional trees where the leaf nodes are local minima and other nodes represent decision points such as saddle points and points of inflection.

U2 - 10.1109/ICASSP.2000.860161

DO - 10.1109/ICASSP.2000.860161

M3 - Other

SP - 3522

EP - 3525

T2 - IEEE International Conference Acoustics Speech and Signal Processing, ICASSP2000

Y2 - 5 June 2000 through 9 June 2000

ER -