Function Approximator System Based on rbf
A function approximation and base technology, which is applied in the field of function approximation system, can solve the problems of function approximation such as large volume, slow operation speed, and not easy to carry
Active Publication Date: 2018-09-18
FUZHOU UNIV
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The purpose of the present invention is to provide a function approximator system based on RBF, which has the advantages of small size, portability, embeddability, etc., can realize a high degree of parallel computing, and overcomes the large volume, non-portability and difficulty of implementing a function approximator by software. Defects of embedding and slow operation speed
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0038] The technical solution of the present invention will be specifically described below in conjunction with the accompanying drawings.
[0039] Such as Figure 1-2 Shown, a kind of function approximator system based on RBF of the present invention, comprises the first RBF neuron circuit module, the second RBF neuron circuit module and the first to the 3rd Gilbert multiplier; Said first RBF neuron circuit module Including the fourth Gilbert multiplier, the first square root circuit and the first Gaussian circuit connected in sequence, the connection between the first square root circuit and the first Gaussian circuit is connected to GND through the first resistor, and the second The RBF neuron circuit module includes a fifth Gilbert multiplier, a second square root circuit and a second Gaussian circuit connected in sequence, and the connection between the second square root circuit and the second Gaussian circuit is connected to GND through a second resistor , the first in...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to a function approximator system based on a RBF. The function approximator system based on the RBF is designed by utilizing a square root extraction circuit, a Gaussian-like function generation circuit and a Bilbert multiplier and other basic circuit units. The function of function approximation can be realized by giving appropriate external offset voltage. The function approximator system based on the RBF can be integrated into a dedicated neural network chip and has the advantages of being small in size, convenient to carry and embeddable so that high degree of parallel computing can be realized, and the defects of software for realizing the function approximator that the size is large, carrying and embedding are inconvenient and computing speed is slow can be overcome. Besides, function approximator system based on the RBF has great extensibility and can be used for solving the approximation problem of more complex functions so that the function approximator system based on the RBF can be widely applied in the field of function approximation and other artificial intelligence by means of the advantages of embeddability, portability, high speed and extensibility.
Description
technical field [0001] The invention relates to a function approximator system based on RBF. Background technique [0002] The theoretical model of RBF (Radial Basic Function) neural network has been widely used in artificial intelligence fields such as function approximation and pattern classification, but it is still mainly focused on the traditional computer software simulation implementation. The software implementation of RBF neural network uses a general-purpose CPU processor, which is inconvenient to embed into other application systems, and relies on a huge general-purpose computer system to complete learning calculations, which is not portable. During the calculation process, the CPU often waits until the neurons of the RBF are calculated one by one before calculating the total result. The serial calculation method is adopted, and the speed is relatively slow. Therefore, the software implementation of RBF neural network is difficult to meet its high-speed, portable...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/063
CPCG06N3/063
Inventor 魏榕山林汉超刘章旺陈林城
Owner FUZHOU UNIV