Hamiltonian neural nets as a universal signal processor / Wiesław Sienko, Wiesław M. Citko.
Rodzaj materiału:
ArtykułJęzyk: angielski Szczegóły wydania: 2002.
W: IEEE. - 2002Streszczenie: This paper presents how to find an architecture for very large scale lossless neural nets, which can be used as Haar-Walsh spectrum analyzers. This analysis relies on the orthogonality of weight matrices W, where W could be Hurwitz-Radon matrices. The unique featrue of these nets is the possibility to treat them either as algorithms or as Hamiltonian physical objects (Haar-Walsh Signal Processors).
Brak egzemplarzy dla tego rekordu
Dane z Informatora o publikowanych wynikach prac naukowo-badawczych w 2002 roku Wydziału Elektroniki.
This paper presents how to find an architecture for very large scale lossless neural nets, which can be used as Haar-Walsh spectrum analyzers. This analysis relies on the orthogonality of weight matrices W, where W could be Hurwitz-Radon matrices. The unique featrue of these nets is the possibility to treat them either as algorithms or as Hamiltonian physical objects (Haar-Walsh Signal Processors).