Frans van den Bergh  Home/Search
Context Related
| cs.up.ac.za/~fvdbe..._weight_init.ps.gz Cached: PS.gz PS PDF This document uses CoBlitz to cache paper downloads. If your firewall is blocking outgoing connections to port 3125, you can use these links to download local copies.
Image Update HelpPS.gz PS PDF From: cs.up.ac.za/~fvdbe...publications (more) (Enter author homepages) |
Rate this article:      (best)Comment on this article |
Abstract: Many training algorithms (like gradient descent, for example) use random initial weights. These algorithms are rather sensitive to their starting position in the error space, which is represented by their initial weights. This paper shows that the training performance can be improved significantly by using a Particle Swarm Optimizer (PSO) to initialize the weights, rather than random initialization. INTRODUCTION It has been shown that Multi-Layer Perceptron (MLP) networks can be used with... (Update)
Context of citations to this paper: More
...being solved. For the purposes of this study, the tness function is the sum squared error (SSE) The interested reader is also referred to [12] for more information on using PSO to train NNs with summation units. Criteria for optimization Three performance criteria for...
...Review Categories: G.1.6, I.2. 6 1 Introduction Particle Swarm Optimizers (PSOs) have previously been used to train neural networks[6, 10] and generally met with success. The advantage of the PSO over many of the other optimization algorithms is its relative simplicity. This...
Cited by: More
Cooperative Learning in Neural Networks using Particle.. - van den Bergh.. (2000) (Correct)
Training Product Units In Feedforward Neural Networks Using .. - Ismail, Engelbrecht (1999) (Correct)
Active bibliography (related documents): More All
0.0: Training Product Unit Networks using Cooperative.. - van den Bergh.. (2001) (Correct)
0.0: Hybrid Particle Swarm - Evolutionary - Algorithm For Search (Correct)
0.0: Particle Swarm Optimization Method in Multiobjective Problems - Parsopoulos, Vrahatis (2002) (Correct)
Similar documents based on text: More All
1.2: Extending Particle Swarm Opimisers with Self-Organized.. - Løvbjerg, Krink (2002) (Correct)
1.1: Hybird Particle Swarm Optimizer with Mass Extinction - Xie, Zhang, Yang (2002) (Correct)
0.9: A Diversity-Guided Particle Swarm Optimizer - the ARPSO - Riget, Vesterstrøm (2002) (Correct)
Related documents from co-citation: More All
2: Computational Intelligence PC Tools (context) - Eberhart, Simpson et al. - 1996
BibTeX entry: (Update)
F van den Bergh, Particle Swarm Weight Initialization in Multi-layer Perceptron Articial Neural Networks, accepted for ICAI, Durban, South Africa, 1999. http://citeseer.ist.psu.edu/vandenbergh99particle.html More
@misc{ bergh99particle,
author = "F. van den Bergh",
title = "Particle Swarm Weight Initialization in Multi-layer Perceptron Articial
Neural Networks",
text = "F van den Bergh, Particle Swarm Weight Initialization in Multi-layer Perceptron
Articial Neural Networks, accepted for ICAI, Durban, South Africa, 1999.",
year = "1999",
url = "citeseer.ist.psu.edu/vandenbergh99particle.html" }
Citations (may not include all citations):1662 Neural Networks for Pattern Recognition (context) - Bishop - 1995
696 UCI repository of machine learning databases (context) - Blake, Keogh et al. - 1998
56 A scaled conjugate gradient algorithm for fast supervised le.. (context) - Moller - 1993
30 Computational Intelligence PC Tools (context) - Eberhart, Simpson et al. - 1996
21 A new optimizer using particle swarm theory (context) - Eberhart, Kennedy - 1995
Documents on the same site (http://www.cs.up.ac.za/~fvdbergh/publications.html): More
Cooperative Learning in Neural Networks using Particle.. - van den Bergh.. (2000) (Correct)
Effects of Swarm Size on Cooperative Particle Swarm.. - van den Bergh, Engelbrecht (Correct)
Training Product Unit Networks using Cooperative.. - van den Bergh.. (2001) (Correct)
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC