id: 30658
Title: The new basic realizations of operations “equivalence” of neuro-fuzzy and bioinspired neuro-logics to create hardware accelerators of advanced equivalental models of neural structures and machine vision systems
Authors: Krasilenko V., Yurchuk N., Lazarev А.
Keywords: self-learning equivalent-convolutional neural structures, equivalent models, continuous-logical operations, hardware accelerator, bioinspired neuro-logic, neuro-fuzzy logic, neuron-equivalentor, current mirror, sorting node, operations “equivalence” and “
Date of publication: 2022-02-08 21:10:10
Last changes: 2022-02-08 21:10:10
Year of publication: 2021
Summary: The perspective of neural networks equivalental models (EM) base on vector-matrix procedure with basic operations of continuous and neuro-fuzzy logic (equivalence, absolute difference) are shown. Capacity on base EMs exceeded the amount of neurons in 4-10 times. This is larger than others neural networks paradigms. Amount neurons of this neural networks on base EMs may be 10 – 100 thousand. The base operations in EMs are normalized equivalence operations. The family of new operations “equivalence” and “non-equivalence” of neuro-fuzzy logic’s, which we have elaborated on the based of such generalized operations of fuzzy-logic’s as fuzzy negation, t-norm and s-norm are shown. Generalized rules of construction of new functions (operations) “equivalence” which uses operations of t-norm and s-norm to fuzzy negation are proposed. Despite the wide variety of types of operations on fuzzy sets and fuzzy relations and the related variety of new synthesized equivalence operations based on them, it is possible and necessary to select basic operations, taking into account their functional completeness in the corresponding algebras of continuous logic, as well as their most effective circuitry implementations.
URI: http://socrates.vsau.edu.ua/repository/getfile.php/30658.pdf
Publication type: Статті у наукових фахових виданнях України (Copernicus та інші)
Publication: Вісник Хмельницького національного університету. Серія: Технічні науки. 2021. № 6 (303). С. 153-166.
In the collections :
Published by: Адміністратор
File : 30658.pdf Size : 1281897 byte Format : Adobe PDF Access : For all
| |
|
|
