论文信息 - Kernel-based topographic map formation achieved with normalized Gaussian competition

Kernel-based topographic map formation achieved with normalized Gaussian competition

A new learning algorithm for kernel-based topographic map formation is introduced. The kernels are Gaussians, and their centers and ranges individually adapted so as to yield an equiprobabilistic topographic map. The converged map also generates a heteroscedastic Gaussian mixture model of the input density. This is verified for both synthetic and real-world examples, and compared with other algorithms for kernel-based topographic map formation.

Marc M. Van Hulle

[1] Akio Utsugi. Hyperparameter Selection for Self-Organizing Maps , 1997, Neural Computation.

[2] K. Obermayer,et al. PHASE TRANSITIONS IN STOCHASTIC SELF-ORGANIZING MAPS , 1997 .

[3] Marc M. Van Hulle. Joint Entropy Maximization in Kernel-Based Topographic Maps , 2002, Neural Computation.

[4] Richard Durbin,et al. An analogue approach to the travelling salesman problem using an elastic net method , 1987, Nature.

[5] M. Hulle. Kernel-Based Equiprobabilistic Topographic Map Formation , 1998, Neural Computation.

[6] Marc M. Van Hulle. Kernel-Based Topographic Map Formation by Local Density Modeling , 2002, Neural Computation.

[7] Jouko Lampinen,et al. On the generative probability density model in the self-organizing map , 2002, Neurocomputing.

[8] Teuvo Kohonen,et al. Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[9] Hujun Yin,et al. Self-organizing mixture networks for probability density estimation , 2001, IEEE Trans. Neural Networks.