For the Self Organizing Map, what is the default neighborhood function?

2 次查看（过去 30 天）
evanma
evanma 2017-8-19
编辑： KSSV 2019-4-3
When it comes to the self organizing map, for example the way it is done using the selforgmap function, what is the default neighborhood radius function and learning restraint? I had been digging through the functions that comprise the self organizing map function and was not sure that I could tell where they are. When there is a target weight vector W(q) for iteration q, we can start with the weight updating function W(q) = W(q-1) + N(q,v,s)*a(s)*(D(t) - W(q-1)) where t is the index of a given input weight vector, D(t) are the input vector's weights, v is the index of a map vector and s is the index of the BMU. Naturally, N(q,v,s) is the neighborhood function and a(s) is the learning restraint.
For the matlab function selforgmap, what is the default function used for N(q,v,s) and a(s)? I was looking to see where these functions are defined in the selforgmap code but was not sure. What are the default functions used?

请先登录，再回答此问题。

回答（1 个）

gunjan
gunjan 2018-4-24
编辑：KSSV 2019-4-3
https://www.mathworks.com/help/nnet/ug/cluster-with-self-organizing-map-neural-network.html#bss4b_l-3
check this.. it says that the weights are updated : W(k+1) = W(k) + alpha * [ (dist) ] rule.. where dist = distance between current neuron that you are updating and the winning neuron. k = iteration number alpha = learning rate

类别

在 Help Center 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by