Metadata language
Książka = Book ; KS/1/2002/P10
Creator: Publisher:Instytut Badań Systemowych. Polska Akademia Nauk ; Systems Research Institute. Polish Academy of Sciences
Place of publishing: Date issued/created: Description:[1-2], 153-165 pages ; 21 cm ; Bibliography p. 164-165
Type of object: Abstract:In the artificial neural networks literature Little attention has been given to consideration of neural networks from the point of view of graph theory. Examination of neural networks as flow graphs gives very interesting and new properties of the neural networks' learning process. The approach is based on tools used in electric circuits when Kirchhoff s laws may not be valid. The analysis leads to the proper equations of the backpropagation algorithm, but in a much simpler manner. The graph methodology incorporates the reciprocal graphs in which signals flow in opposite directions. Neural networks in which signals flow in opposite direction are called the adjoint neural networks.
Relation: Resource type: Detailed Resource Type: Source: Language: Language of abstract: Rights:Creative Commons Attribution BY 4.0 license
Terms of use:Copyright-protected material. [CC BY 4.0] May be used within the scope specified in Creative Commons Attribution BY 4.0 license, full text available at: ; -
Digitizing institution:Systems Research Institute of the Polish Academy of Sciences
Original in:Library of Systems Research Institute PAS
Projects co-financed by: Access: