R. Kamimura, O. Uchida, and S. Hashimoto (Japan)
: mutual information maximization, competi
tive learning, winner-take-all, Minkowski distance
In this paper, we propose a new network-growing method
to accelerate learning and to extract explicit features in
complex input patterns. We have so far proposed a new
type of network-growing algorithm called greedy network
growing algorithm,. Though the method have shown
some potentiality to extract salient features, we have ob
served that the method is slow in learning, and sometimes
it cannot produce a state where information is large enough
to produce explicit internal representations. To remedy
this shortcoming, we introduce here Minkowski distance
between input patterns and connection weights used to
produce competitive unit outputs. When the parameter for
Minkowski distance is larger, some detailed parts in input
patterns can be eliminated, which enables networks to
converge faster and to extract main parts of input patterns.
We applied our new method to an economic data analysis.
Experimental results confirm that a new method with
Minkowski distance can significantly accelerate learning,
and clearer features can be extracted.