日本語
 
User Manual Privacy Policy ポリシー/免責事項 連絡先
  詳細検索ブラウズ

アイテム詳細


公開

報告書

View-based cognitive mapping and path planning

MPS-Authors
/persons/resource/persons84193

Schölkopf,  B
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

/persons/resource/persons84072

Mallot,  HA
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

URL
There are no locators available
フルテキスト (公開)

MPIK-TR-7.pdf
(出版社版), 404KB

付随資料 (公開)
There is no public supplementary material available
引用

Schölkopf, B., & Mallot, H.(1994). View-based cognitive mapping and path planning (7). Tübingen, Germany: Max Planck Institute for Biological Cybernetics.


引用: http://hdl.handle.net/11858/00-001M-0000-0013-ED18-9
要旨
We present a scheme for learning a cognitive map of a maze from a sequence of views and movement decisions. The scheme is based on an intermediate representation called the view graph. We show that this representation carries sufficient information to reconstruct the topological and directional structure of the maze. Moreover, we present a neural network that learns the view graph during a random exploration of the maze. We use a unsupervised competitive learning rule which translates temporal sequence (rather than similarity) of views into connectedness in the network. The network uses its knowledge of the topological and directional structure of the maze to generate expectations about which views are likely to be perceived next, improving the view recognition performance. We provide an additional mechanism which uses the map to find paths between arbitrary points of the previously explored environment. The results are compared to findings of behavioural neuroscience.