PDE Seminar (2018/10/26): 適応フィルタが創発する自己組織化マップとそのデモンストレーション
 Date

20181026 16:30 
20181026 18:00
 Place
 Faculty of Science Building #3, Room 309

Speaker/Organizer
 Hideo Kubo (Hokkaido University), Kogo Yoshikawa (Hokkaido University)

 For the modeling of the brain, the adaptive filter is introduced by Kohonen. It is a frame allowing for elementary operation to change with time, and this is described by the differential equation basically. Furthermore, the order by a certain measure among input data can be almost preserved about the order by other measure among output data, if we introduce mechanism of the lateral interaction into adaptive filters. The order to occur in this way is known as SelfOrganizing Map (SOM). We will show a computer demo playing SOM software with a toy problem. The software implementation of SOM represents the result of a vector quantization algorithm that places a number of reference or codebook vectors into a highdimensional input data space to approximate to its data sets in an ordered fashion. This demo would help us to understand why and how SOM works.