Hyperparameter selection for self-organizing maps
Neural Computation
Self-organizing maps
GTM: the generative topographic mapping
Neural Computation
Issues in Bayesian analysis of neural network models
Neural Computation
Neural maps and topographic vector quantization
Neural Networks
Bayesian Sampling and Ensemble Learning in Generative Topographic Mapping
Neural Processing Letters
Training Kohonen Feature Maps in Different Topologies: An Analysis Using Genetic Algorithms
Proceedings of the 5th International Conference on Genetic Algorithms
A nonlinear projection method based on Kohonen's topology preserving maps
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Bayesian inference for neural networks has received a good deal of attention in recent years. Unlike standard methods, the bayesian approach provides the analyst with the richness (and complexity) of a probability distribution over the space of network weights (and possibly other quantities of interest). These posterior distributions prompt an optimization problem that may be suitable for evolutionary algorithms. This possibility is obviously of foremost interest when no alternative global functions are available for optimization. Some preliminary results related to one of such cases, namely, the self-organizing map, are presented in this paper. Specifically, a familiar "Steady-state" diffusion genetic algorithm is described and tested.