混合SOM(带有MLP) [英] Hybrid SOM (with MLP)

问题描述

有人可以提供一些信息，说明如何将自组织图与多层感知器正确结合吗?

Could someone please provide some information on how to properly combine a self organizing map with a multilayer perceptron?

我最近阅读了一些与常规MLP相比的有关此技术的文章，并且该方法在预测任务中的表现更好.因此，我想通过对输入数据进行聚类并将结果传递到MLP后端，将SOM用作减少维度的前端.

I recently read some articles about this technique in comparison to regular MLPs and it performed way better in prediction tasks. So, I want to use the SOM as front-end for dimension reduction by clustering the input data and pass the results to an MLP back-end.

我目前实现它的想法是用几个训练集训练SOM并确定集群.之后，我使用与SOM群集一样多的输入单元来初始化MLP.下一步将是使用SOM的输出(哪个值?... BMU的权重?)作为网络的输入(群集与输入单元匹配的SOM的输出，其他任何输入单元的零?)来训练MLP.

My current idea of implementing it is it to train the SOM with a couple of training sets and to determine the clusters. Afterwards, I initialize the MLP with as many input units as SOM clusters. Next step would be to train the MLP using the SOM's output (which value?...weights of BMU?) as in input for the network (SOM's Output for the Cluster matching Input Unit and zeros for any other Input Units?).

推荐答案

没有唯一的方法可以做到这一点.让我列出一些可能性:

There is no single way of doing that. Let me list some possibilities:

• 您描述的那个.但是，您的MLP将需要具有K * D输入，其中K是集群数，D是输入维.没有降维.
• 与您的想法类似，但不使用权重，只需将1发送给BMU，将0发送给其余集群.然后，您的MLP将需要K个输入.
• 与上面相同，但不是1或0，而是将输入向量到每个聚类的距离发送出去.
• 与上述相同，但要代替距离，而是为每个群集计算一个高斯激活.
• 由于SOM保留拓扑，因此仅发送BMU的2D坐标(可能在0到1之间标准化).那么您的MLP仅需要2个输入，就可以实现真正的极端降维.

您可以在此处了解有关这些想法的更多信息: SOM的主要时间扩展:概述.它不是将SOM的输出馈送到MLP，而是将SOM馈送到其自身.但是，当您尝试从SOM生成一些输出时，您将能够理解各种可能性.

You can read about those ideas and some more here: Principal temporal extensions of SOM: Overview. It is not about feeding the output of a SOM to a MLP, but a SOM to itself. But you'll be able to understand the various possibilities when trying to produce some output from a SOM.

这篇关于混合SOM(带有MLP)的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！