Unlock instant, AI-driven research and patent intelligence for your innovation.

Data processing processor, corresponding method and computer program

A computer program and data processing technology, applied in the field of neural network materialization, can solve the problem of not solving the problem of hardware implementation of neural network, and achieve the effect of reducing computing load

Inactive Publication Date: 2021-08-17
UNIV DE BOURGOGNE (FR) +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0016] However, the solutions described in these articles do not address the hardware implementation of general neural networks (i.e. neural networks implementing general neurons), which can implement multiple neural network types, including several activation functions and / or a hybrid neural network of several combination functions

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Data processing processor, corresponding method and computer program
  • Data processing processor, corresponding method and computer program
  • Data processing processor, corresponding method and computer program

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0068] 5.1. Description of technical principles

[0069] 5.1.1. overview

[0070] Faced with the problem of implementing an adaptive and configurable neural network, the inventors focused on the materialization of computation implemented in different configurations. As explained above, the difference between neural networks is mainly in the calculations performed. Specifically, the layers that make up a neural network implement individual neurons that perform combination and activation functions, which may vary from network to network. Today, on a given electronic device (such as a smartphone, tablet, or personal computer), many different neural networks can be implemented, each used by a different application or process. Therefore, in order to effectively implement such neural networks, it is not possible to have dedicated hardware components for each type of neural network to be implemented. For this reason, most neural networks today are implemented purely in softwar...

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

No PUM Login to View More

Abstract

The invention relates to a data processing processor, said processor comprising at least one processing memory (MEM) and one computation unit (CU). According to the invention, the computation unit (CU) comprises a set of configurable computation units called configurable neurones, each configurable neurone (CN) of the set of configurable neurones (SCN) comprising a module for computing combination functions (MCCF) and a module for computing activation functions (MCAF), each module for computing activation functions (AFU) comprising a register for receiving a configuration command, so that said command determines an activation function to be executed from at least two activation functions that can be executed by the module for computing activation functions (AFU).

Description

technical field [0001] The present invention relates to the materialization of neural networks. More specifically, the present invention relates to physical implementations of adaptive and configurable neural networks. Still more specifically, the present invention relates to embodiments of general neural networks whose configuration and operation can be adjusted as desired. Background technique [0002] In the field of computerized data processing, a neural network is a digital system whose design was originally inspired by the function of biological neurons. Neural networks are more generally modeled as systems that include processing algorithms and statistics (including weights). Processing algorithms allow input data to be processed and combined with statistical data to obtain output results. A processing algorithm consists of defining a computation that is performed on the input data as well as the statistics of the network to provide an output result. Meanwhile, co...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/04G06N3/063
CPCG06N3/063G06N3/048
Inventor 米歇尔·杜索米歇尔·潘达文
Owner UNIV DE BOURGOGNE (FR)