110 likes | 264 Views
Machine fusion to enhance the topology preservation of vector quantization artificial neural networks. R. Salas, C. Saavedra , H. Allende, C. Moraga PRL, 2011 Presented by Hung-Yi Cai 2011/5/25. Outlines. Motivation Objectives Methodology Experiments Conclusions Comments.
E N D
Machine fusion to enhance the topology preservation of vector quantization artificial neural networks R. Salas, C. Saavedra, H. Allende, C. Moraga PRL, 2011 Presented by Hung-Yi Cai 2011/5/25
Outlines • Motivation • Objectives • Methodology • Experiments • Conclusions • Comments
Motivation • The objective of VQ is to preserve the topological relationships existing in a data set and to project the data to a lattice of lower dimensions. • It’s difficult to properly specify the structure of the lattice that best preserves the topology of the data.
Objectives • Bagging • Boosting To propose a merging algorithms for machine-fusion, boosting-fusion-based and hybrid-fusion ensembles of SOM, NG and GSOM networks.
Methodology • Machine fusion method for the ensemble of VQ-ANN
Methodology • Machine fusion method for the ensemble of VQ-ANN
Methodology • Boosting machine fusion method
Experiments Synthetic Data
Experiments Real Data
Conclusions • The main goal of this paper is to improve the topology preservationby combining the output of several VQ-ANN. • The proposed ensemble schemes were able to improve the quality of topological representation compared to their respective base single networks.
Comments • Advantages • Improving the VQ in the ANN. • Drawbacks • The methods don’t consider the outliers. • Applications • Vector Quantization