Scatter-Partitioning RBF Network for Function Regression and Image Segmentation: Preliminary Results

TitleScatter-Partitioning RBF Network for Function Regression and Image Segmentation: Preliminary Results
Publication TypeTechnical Report
Year of Publication1998
AuthorsBaraldi, A.
Other Numbers1138
KeywordsGestaltist theory, image segmentation, low-level vision, prototype vectors, RBF networks, supervised and unsupervised learning from data, synaptic links

Scatter-partitioning Radial Basis Function (RBF) networks increase their number of degrees of freedom with the complexity of an input-output mapping to be estimated on the basis of a supervised training data set. Due to its superior expressive power a scatter-partitioning Gaussian RBF (GRBF) model, termed Supervised Growing Neural Gas (SGNG), is selected from the literature. SGNG employs a one-stage error-driven learning strategy and is capable of generating and removing both hidden units and synaptic connections. A slightly modified SGNG version is tested as a function estimator when the training surface to be fitted is an image, i.e., a 2-D signal whose size is finite. The relationship between the generation, by the learning system, of disjointed maps of hidden units and the presence, in the image, of pictorially homogeneous subsets (segments) is investigated. Unfortunately, the examined SGNG version performs poorly both as function estimator and image segmenter. This may be due to an intrinsic inadequacy of the one-stage error-driven learning strategy to adjust structural parameters and output weights simultaneously but consistently. In the framework of RBF networks, further studies should investigate the combination of two-stage error-driven learning strategies with synapse generation and removal criteria.

Bibliographic Notes

ICSI Technical Report TR-98-017

Abbreviated Authors

A. Baraldi

ICSI Publication Type

Technical Report