Dynamic tree topology learning by self-organization

Abstract:

The discovery of the underlying topology of real-world data is a difficult task due to the high-dimensional and the complex structure in real datasets. In some cases, when the topology of the data is not known or the information is provided in a stream, it is advantageous to learn tree topologies from the data. This task can be carried out by dynamic self-organizing neural networks, so that the specific topology of the dataset is discovered by the network. In this work a self-organizing spanning tree is proposed, which is able to learn a tree topology without any prespecified structure. Experimental results are provided to show the performance of the model with real video data for a foreground detection task. Comparative results are reported.

Año de publicación:

2017

Keywords:

• Computer Vision
• Spanning trees
• unsupervised learning
• Self-organizing map topologies

Fuente:

scopus