论文信息 - Incremental Learning of Local Linear Mappings

Incremental Learning of Local Linear Mappings

A new incremental network model for supervised learning is proposed. The model builds up a structure of units each of which has an associated local linear mapping (LLM). Error information obtained during training is used to determine where to insert new units whose LLMs are interpolated from their neighbors. Simulation results for several classiication tasks indicate fast convergence as well as good generalization. The ability of the model to also perform function approximation is demonstrated by an example.

Bernd Fritzke

[1] Bernd Fritzke,et al. Growing cell structures--A self-organizing network for unsupervised and supervised learning , 1994, Neural Networks.

[2] Helge Ritter,et al. Cascade LLM Networks , 1992 .

[3] Bernd Fritzke,et al. A Growing Neural Gas Network Learns Topologies , 1994, NIPS.

[4] Klaus Schulten,et al. Topology-conserving maps for learning visuo-motor-coordination , 1989, Neural Networks.

[5] T. Martínez,et al. Competitive Hebbian Learning Rule Forms Perfectly Topology Preserving Maps , 1993 .

[6] Thomas Martinetz,et al. 'Neural-gas' network for vector quantization and its application to time-series prediction , 1993, IEEE Trans. Neural Networks.