Reducing the number of neurons in radial basis function networks with dynamic decay adjustment

Paetz, Jürgen

doi:10.1016/j.neucom.2003.12.004

Cited by 39 publications

(22 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As such, FAMDDA-HGA was selected for further performance comparison with other classification models. These include two constructive DDA-based ANNs, i.e., RBFN-DDA and its improved version with a pruning algorithm, RBFN-DDA-T (Paetz 2004); a swarm intelligence-based model, i.e., Self-organising swarm (SOSwarm) (O'Neill and Brabazon 2008), which utilises an adaptive particle swarm algorithm with a history mechanism and cognitive learning for refining the weights of a self-organising map network; and memeticbased models, i.e., memetic evolutionary algorithm-micro GA (EMA-lGA) and memetic evolutionary algorithmartificial immune system (EMA-AIS) (Ang et al 2010), which combine both global and local search algorithms for performing data classification and rule extraction.…”

Section: Comparison With Other Classification Modelsmentioning

confidence: 99%

“…To compare with the RBFN-DDA and RBFN-DDA-T networks as in Paetz (2004), the PID, AUS, and HEA data sets were used. Each data set was divided equally for training and test, and the experiment was repeated eight times.…”

Section: Comparison With Other Classification Modelsmentioning

confidence: 99%

“…4. For comparison purposes, the results of RBFN-DDA, RBFN-DDA-T (Paetz 2004), SOSwarm (O'Neill and Brabazon 2008), EMA-lGA and EMA-AIS (Ang et al 2010) are also shown in Table 4 (for numerical comparison) and Fig. 4 (for graphical comparison).…”

Section: Comparison With Other Classification Modelsmentioning

confidence: 99%

See 2 more Smart Citations

Integration of supervised ART-based neural networks with a hybrid genetic algorithm

Tan

Lim

2010

Soft Comput

View full text Add to dashboard Cite

In this paper, two evolutionary artificial neural network (EANN) models that are based on integration of two supervised adaptive resonance theory (ART)-based artificial neural networks with a hybrid genetic algorithm (HGA) are proposed. The search process of the proposed EANN models is guided by a knowledge base established by ART with respect to the training data samples. The EANN models explore the search space for ''coarse'' solutions, and such solutions are then refined using the local search process of the HGA. The performances of the proposed EANN models are evaluated and compared with those from other classifiers using more than ten benchmark data sets. The applicability of the EANN models to a real medical classification task is also demonstrated. The results from the experimental studies demonstrate the effectiveness and usefulness of the proposed EANN models in undertaking pattern classification problems.

show abstract

Section: Comparison With Other Classification Modelsmentioning

confidence: 99%

Section: Comparison With Other Classification Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Integration of supervised ART-based neural networks with a hybrid genetic algorithm

Tan

Lim

2010

Soft Comput

View full text Add to dashboard Cite

show abstract

“…An online pruning technique [21] to ascertain outliers/ superfluous nodes and the network complexity of FAM-DDA/FAM at each training epoch is proposed. The online pruning algorithm participates in the learning process of FAMDDA/FAM.…”

Section: Online Pruning Algorithmmentioning

confidence: 99%

“…The chosen data sets contained various types of input attributes (real, ordered, binary and nominal values) at arbitrary complexities in the data space. The objective of this set of experiments was to compare the performances of FAMDDA/FAMDDA-T and FAM/ FAM-T using the same experimental procedure as have been adopted in [21]. Each data set was divided into 50% (randomly selected samples) for training and 50% for test.…”

Section: The Uci Benchmark Data Setsmentioning

confidence: 99%

An online pruning strategy for supervised ARTMAP-based neural networks

Tan

Rao

Lim

2008

Neural Comput & Applic

View full text Add to dashboard Cite

Identifying an appropriate architecture of an artificial neural network (ANN) for a given task is important because learning and generalisation of an ANN is affected by its structure. In this paper, an online pruning strategy is proposed to participate in the learning process of two constructive networks, i.e. fuzzy ARTMAP (FAM) and fuzzy ARTMAP with dynamic decay adjustment (FAM-DDA), and the resulting hybrid networks are called FAM/ FAMDDA with temporary nodes (i.e. FAM-T and FAM-DDA-T, respectively). FAM-T and FAMDDA-T possess a capability of reducing the network complexity online by removing unrepresentative neurons. The performances of FAM-T and FAMDDA-T are evaluated and compared with those of FAM and FAMDDA using a total of 13 benchmark data sets. To demonstrate the applicability of FAM-T and FAMDDA-T, a real fault detection and diagnosis task in a power plant is tested. The results from both benchmark studies and real-world application show that FAMDDA-T and FAM-T are able to yield satisfactory classification performances, with the advantage of having parsimonious network structures.

show abstract

Supervised Neuro-fuzzy Clustering for Life Science Applications

Paetz

2006

Biological and Medical Data Analysis

View full text Add to dashboard Cite

Reducing the number of neurons in radial basis function networks with dynamic decay adjustment

Cited by 39 publications

References 9 publications

Integration of supervised ART-based neural networks with a hybrid genetic algorithm

Integration of supervised ART-based neural networks with a hybrid genetic algorithm

An online pruning strategy for supervised ARTMAP-based neural networks

Supervised Neuro-fuzzy Clustering for Life Science Applications

Contact Info

Product

Resources

About