论文信息 - Modified growing neural gas algorithm for faster convergence on signal distribution sudden change

Modified growing neural gas algorithm for faster convergence on signal distribution sudden change

The paper deals with the problem of faster optimal coverage of a Growing Neural Gas algorithm for random signals appearing with non-stationary distributions. A modification of the algorithm that successfully solves this problem will be presented with simulations in a 2-D environment and statistical results that will show its efficiency. A comparison with a previous solution for the same problem using so called Utility measure will be also given.

Andrea Kulakov | Stojanco Gancev | A. Kulakov | S. Gancev

[1] Krzysztof Ciesielski,et al. Incremental document map formation: multi-stage approach , 2006, Ann. UMCS Informatica.

[2] José Alfredo Ferreira Costa,et al. Cluster Analysis using Growing Neural Gas and Graph Partitioning , 2007, 2007 International Joint Conference on Neural Networks.

[3] Xiang Cao,et al. Video shot motion characterization based on hierarchical overlapped growing neural gas networks , 2003, Multimedia Systems.

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

[5] Bernd Fritzke,et al. A Self-Organizing Network that Can Follow Non-stationary Distributions , 1997, ICANN.

[6] Li Li,et al. Hybrid Learning Algorithm for Effective Coverage in Wireless Sensor Networks , 2008, 2008 Fourth International Conference on Natural Computation.

[7] Li Li,et al. Dynamic Coverage Based on Neural Gas Learning Algorithm for Wireless Sensor Network , 2009 .

[8] Jim Holmström,et al. Examensarbete DV 3 2002-0830 Growing Neural Gas Experiments with GNG , GNG with Utility and Supervised GNG , 2002 .