CN107563499A - A kind of design method of the more threshold values polygamma function artificial neurons of codified - Google Patents
A kind of design method of the more threshold values polygamma function artificial neurons of codified Download PDFInfo
- Publication number
- CN107563499A CN107563499A CN201710809304.6A CN201710809304A CN107563499A CN 107563499 A CN107563499 A CN 107563499A CN 201710809304 A CN201710809304 A CN 201710809304A CN 107563499 A CN107563499 A CN 107563499A
- Authority
- CN
- China
- Prior art keywords
- threshold values
- codified
- artificial neuron
- value
- input
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of technical field of the design method of the more threshold values polygamma function artificial neurons of codified, it is to belong to artificial intelligence, bionics, the technical field of circuit design, major technique is that artificial neuron is inputted by multichannel, when accumulated value is less than minimum threshold values, artificial neuron, it will not be activated, when cumulative value exceedes the threshold values of setting, artificial neuron is activated, artificial neuron is provided with multiple threshold values, according to cumulative value, reach that threshold values, activation primitive with regard to starting that corresponding threshold values, value is passed to codified and selects end-apparatus, codified selects setting of the end-apparatus according to control terminal, those ports are opened in selection, those circuit outputs, those circuits are closed.
Description
Technical field
A kind of technical field of the design method of the more threshold values polygamma function artificial neurons of codified, is to belong to artificial intelligence,
Bionics, the technical field of circuit design, major technique are that artificial neuron is inputted by multichannel, when accumulated value is less than minimum valve
During value, artificial neuron, it will not be activated, when cumulative value exceedes the threshold values of setting, artificial neuron is activated, artificial neuron
Member is provided with multiple threshold values, according to cumulative value, reaches that threshold values, just starts the activation primitive of that corresponding threshold values, value is passed
Pass codified and select end-apparatus, codified selects setting of the end-apparatus according to control terminal, selects those ports to open, those circuits are defeated
Go out, those circuits are closed.
Background technology
Neuron is the elementary cell for forming brain, and the brain of the mankind is that have thousands of individual neurons according to certain rule
Form, for the mankind in order to simulate human brain, the design to artificial neuron is the most important thing, has artificial neuron to form people
Work network, artificial neural network are a kind of mathematical modulos for the structure progress information processing that application is similar to cerebral nerve cynapse connection
Type.In this model, composition network is coupled to each other between substantial amounts of artificial neuron, i.e. " neutral net ", to reach processing
The purpose of information.A kind of kinetic simulation for the distributed parallel information processing algorithm structure for imitating animal nerve network behavior feature
Type., with multichannel input stimulus are received, the part that " excitement " output is produced when exceeding certain threshold value by weighted sum is dynamic to imitate for it
The working method of thing neuron, and the weight coefficient of the structure being coupled to each other by these neural components and reflection strength of association makes
Its " collective behavior " has the various complicated information processing functions.Particularly it is this macroscopically have robust, it is fault-tolerant, anti-interference,
The component performance that is formed as of the flexible and strong function such as adaptability, self study is updated, and passes through the mutual of complexity
Connection relation is achieved, thus artificial neural network is a kind of connection mechanism model, has many key characters of complication system.
Artificial neural network is applied to signal transacting, data compression, pattern-recognition, robot vision, knowledge processing and its application, in advance
Survey, evaluation and the combinatorial optimization problem such as decision problem, scheduling, route planning.It can be used for mould in Control System Design
Intend controlled device characteristic, search and study control law, realize fuzzy and intelligent control, therefore the design to neuron is very
Important, because fairly obvious, the shape of neuron is very more, although the mankind classify it, neuron has on thousands of
Ten thousand kinds, therefore different neurons also possesses different functions, the present invention is the design method of one of which neuron, can be real
The more threshold values polygamma function functions of existing codified, the design very simple of existing neuron, are exactly all inputs and weight phase
Multiply, then added up, subtract threshold values, then activation primitive is set, pass to next layer of neuron.
The content of the invention
The brain of people is that many neurons are formed, therefore neuron is the elementary cell of neutral net, fairly obvious, nerve
First enormous amount, just there are the neuron of different shape, structure, physiologic character and function, neuron in the different parts of human body
Shape it is very strange very more, although the mankind classify to it, neuron has millions upon millions of kinds, therefore different nerves
Member also possesses different functions, and the present invention is that a kind of neuron therein is designed, due to the design of existing neuron
Very simple is single, and exactly all inputs and multiplied by weight are added up, and subtracts threshold values, then sets activation primitive, passes
Pass next layer of neuron, so form a network, and so simple design solve many forefathers of the mankind can not
Solve the problems, such as, tremendous influence, but a kind of this artificial neuron meta structure simply most simply, real generation are produced to All Around The World
The various shapes of neuron in boundary, various functions, therefore will invention various functions neuron design, this hair
Bright is exactly the design method of one of similar a variety of neuronal functions, a kind of more threshold values polygamma function artificial neurons of codified
The design method of member, it is characterized in that：The more threshold values polygamma function artificial neurons of codified are by input, artificial neuron, transmission
Line, codified select end-apparatus, output end composition, and input receives the defeated of upper level artificial neuron such as the input of neuron
Enter or the input by other equipment, the effect of artificial neuron is added up after value and multiplied by weight input, if tired
The value added is less than minimum threshold values, then artificial neuron would not be activated, without any reaction, if cumulative value is more than most
Small threshold values, then artificial neuron is activated, and multiple threshold values are also set up on this minimum threshold values, when cumulative value is more than some
Threshold values, just start the corresponding activation primitive of this threshold values, it is defeated because the activation primitive of setting is different, therefore after being activated
It is also different to go out the information with corresponding activation primitive, and the effect of transmission line is exactly to transmit information, and information can be passed to
The place to compare farther out, as the aixs cylinder of neuron, the effect that codified selects end-apparatus is to receive the letter that activation primitive transmits
Breath, and it is which function passes comes that can be offered an explanation according to information, sets this function to be exported from which port, controller
Effect can change the setting that codified selects end-apparatus, change the port of function output, the effect of output end is exactly various activation
The numerical value of function output is delivered to next layer of artificial neuron, and can be multiplied with weight, once so cumulative value surpasses
That threshold values is crossed, just excites function corresponding to this threshold values, by setting control terminal, allows this function from which port output, wherein
Artificial neuron is made up of using following design, artificial neuron 3 parts, and 1 is accumulator, and 2 be different threshold values, and 3 be different
Activation primitive, the effect of accumulator is added up after input and multiplied by weight last layer, and the design of different threshold values is
It is such, set minimum threshold values a, a<b<c<D, when the value of input is less than a, then artificial neuron would not be activated, if
The value of input is more than a, and artificial neuron is just activated, and the value at this moment inputted will be compared with different threshold values, such as defeated
The value entered is less than d more than c, then and it will start activation primitive f (x3) corresponding to c threshold values, export information corresponding to f (x3), its
Middle codified selects the information for the function that is activated that end-apparatus is transmitted using following design, reception activation primitive collection, and codified is selected
End-apparatus is according to the information transmitted, it is known that is which function passes comes, and knows intensity level, thus according to the inside
Set, the value of this activation primitive, set from which port output, wherein control terminal is designed so as to, and it can change
Codified selects the setting of end-apparatus, for example to correspond to output port be i-1.i-5.i-9 to corresponding f (x3) activation primitive, after setting
Port can be changed to 1-2.i-3.i-4.i-5.i-6.i-7.i-8, this can be designed according to design requirement.
Brief description of the drawings
Fig. 1 is the structure principle chart of the more threshold values polygamma function artificial neurons of codified, i-1.1-2.i-3.i-4.i-5.i-
6.i-7.i-8.i-9.i-10.i-11.i-12 represents input, and this input is a lot, and it is for masterpiece to draw 12 here
With o-1.o-2.o-3.o-4.o-5.o-6.o-7.o-8.o-9.o-10.o-11.o-12 represents output end, and this output end is very
More, it is for role of delegate to draw 12 here, and a-1 represents artificial neuron, and a-2 represents the insideAccumulator, a.b.c.d are
Different threshold values is represented, a-3 represents activation primitive collection, and f (x1) .f (x2) .f (x3) .f (x4) represents different activation primitives, this
In the how many individual functions of the how many individual function sets of needs, this 4 are roles of delegate, according to the threshold values of input, letter corresponding to activation
Number, b-1 codifieds select end-apparatus, and b-2 represents transmission line, and r-1 represents control terminal.
Implementation
After some neurons are activated, it is information by axonal transport to end, but end is not to be activated to be delivered to down entirely
One layer of neuron, be selectively to pass to next layer of neuron, therefore present invention employs such design method, input
End receives the input of upper level artificial neuron or the input by other equipment, artificial neuron such as the input of neuron
Effect be to be added up after value and multiplied by weight input, if cumulative value is less than minimum threshold values, then artificial neuron
Member would not be activated, without any reaction, if cumulative value is more than minimum threshold values, then and artificial neuron is activated,
Multiple threshold values are also set up above this minimum threshold values, when cumulative value is more than some threshold values, just start this threshold values corresponds to activation
Function, because the activation primitive of setting is different, therefore after being activated, the information with respective function password and intensity is exported,
The effect of transmission line is exactly to transmit information, and information can be passed to the place to compare farther out, as the aixs cylinder of neuron, is divided
The effect for sending out device is to receive the information with password and intensity, this information is transmitted to the outlet line of corresponding password, and this is defeated certainly
It is multiple to go out end, and the effect of output end is exactly that the numerical value of various activation primitives output is delivered to next layer of artificial neuron,
And can be multiplied with weight, the neuron as is networked, and it can individually network or artificial with other functions
Neuron networks, and forms an artificial brain, it is possible to reaches the function of imitating human brain, due to the artificial neuron of the present invention
Member, it is using the form of more threshold values polygamma functions orientation, therefore more functions can be realized, can be with fewer of the invention
Artificial neuron, reach sufficiently complex network function.
Claims (1)
1. a kind of design method of the more threshold values polygamma function artificial neurons of codified, it is characterized in that：The more threshold values polygamma functions of codified
Artificial neuron selects end-apparatus by input, artificial neuron, transmission line, codified, output end forms, and input is such as nerve
The input of member, the input of upper level artificial neuron or the input by other equipment are received, the effect of artificial neuron is handle
Added up after the value and multiplied by weight of input, if cumulative value is less than minimum threshold values, then artificial neuron would not be by
Activation, without any reaction, if cumulative value is more than minimum threshold values, then artificial neuron is activated, in this minimum threshold values
Multiple threshold values are also set up above, when cumulative value is more than some threshold values, just start the corresponding activation primitive of this threshold values, due to setting
The activation primitive put is different, therefore after being activated, the information of output band correspondence activation primitive be also it is different, transmission line
Effect is exactly to transmit information, and information can be passed to the place to compare farther out, and as the aixs cylinder of neuron, codified selects end
The effect of device is to receive the information that activation primitive transmits, and it is which function passes comes that can be offered an explanation according to information,
This function is set to change the setting that codified selects end-apparatus from which port output, the effect of controller, it is defeated to change function
The port gone out, the effect of output end are exactly that the numerical value of various activation primitives output is delivered to next layer of artificial neuron, and can
To be multiplied with weight, once so cumulative value exceedes that threshold values, just excite this threshold values corresponding to function, pass through setting
Control terminal, this function is allowed to be designed from which port output, wherein artificial neuron using following, artificial neuron is by 3 part structures
Into 1 is accumulator, and 2 be different threshold values, and 3 be different activation primitives, and the effect of accumulator is input and power last layer
Heavy phase is added up after multiplying, and the design of different threshold values is such, sets minimum threshold values a, a<b<c<D, when the value of input is less than
A, then artificial neuron would not be activated, if the value of input is more than a, artificial neuron is just activated, at this moment inputted
Value will be compared with different threshold values, for example the value inputted is less than d more than c, then will be started and be swashed corresponding to c threshold values
Function f (x3) living, wherein information corresponding to output f (x3), codified select end-apparatus using following design, receive activation primitive collection and pass
The information of the defeated function that is activated to come, codified select end-apparatus according to the information transmitted, it is known that are which function passes mistakes
Come, and know intensity level, thus according to the setting of the inside, the value of this activation primitive, set from which port output,
Wherein control terminal is designed so as to, and it can change the setting that codified selects end-apparatus, such as corresponding f (x3) activation primitive pair
It is i-1.i-5.i-9 to answer output port, and port 1-2.i-3.i-4.i-5.i-6.i-7.i-8 can be changed to after setting, this
It can be designed according to design requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710809304.6A CN107563499A (en) | 2017-09-10 | 2017-09-10 | A kind of design method of the more threshold values polygamma function artificial neurons of codified |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710809304.6A CN107563499A (en) | 2017-09-10 | 2017-09-10 | A kind of design method of the more threshold values polygamma function artificial neurons of codified |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107563499A true CN107563499A (en) | 2018-01-09 |
Family
ID=60980368
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710809304.6A Pending CN107563499A (en) | 2017-09-10 | 2017-09-10 | A kind of design method of the more threshold values polygamma function artificial neurons of codified |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107563499A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103455843A (en) * | 2013-08-16 | 2013-12-18 | 华中科技大学 | Feedback artificial neural network training method and feedback artificial neural network calculating system |
| CN106126481A (en) * | 2016-06-29 | 2016-11-16 | 华为技术有限公司 | A kind of computing engines and electronic equipment |
-
2017
- 2017-09-10 CN CN201710809304.6A patent/CN107563499A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103455843A (en) * | 2013-08-16 | 2013-12-18 | 华中科技大学 | Feedback artificial neural network training method and feedback artificial neural network calculating system |
| CN106126481A (en) * | 2016-06-29 | 2016-11-16 | 华为技术有限公司 | A kind of computing engines and electronic equipment |
Non-Patent Citations (4)
| Title |
|---|
| SAHIL ABROL ET AL.: "Implementation of Single Artificial Neuron Using various Activation Functions and XOR Gate on FPGA chip", 《2015 SECOND INTERNATIONAL CONFERENCE ON ADVANCES IN COMPUTING AND COMMUNICATION ENGINEERING》 * |
| 姚茂群 等: "多阈值神经元电路设计及在多值逻辑中的应用", 《计算机学报》 * |
| 陈允平 等: "《人工神经网络原理及其应用》", 31 August 2002 * |
| 韦一 等: "基于多阈值神经元的D型触发器设计", 《浙江大学学报(理学版)》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105095966B (en) | The hybrid system of artificial neural network and impulsive neural networks | |
| CN108090658A (en) | Arc fault diagnostic method based on time domain charactreristic parameter fusion | |
| CN110073370A (en) | Low power architecture for sparse neural network | |
| CN108197773A (en) | Methods of electric load forecasting, load forecast device and terminal device | |
| CN107480782A (en) | Learn neural network processor on a kind of piece | |
| CN106982359A (en) | A kind of binocular video monitoring method, system and computer-readable recording medium | |
| CN109711534A (en) | Dimensionality reduction model training method, device and electronic equipment | |
| CN111401547B (en) | HTM design method based on circulation learning unit for passenger flow analysis | |
| Wu et al. | A hybrid optimization method of particle swarm optimization and cultural algorithm | |
| CN107563499A (en) | A kind of design method of the more threshold values polygamma function artificial neurons of codified | |
| CN105373829B (en) | A kind of full Connection Neural Network structure | |
| CN107516130A (en) | A kind of more threshold values polygamma functions select the design method of end output artificial neuron | |
| CN107886163A (en) | Single-object problem optimization method and device based on AGN and CNN | |
| CN107563505A (en) | A kind of design method of external control implantation feedback artificial neuron | |
| CN107563506A (en) | A kind of voltage-frequency formula selects the design method that frequency exports artificial neuron | |
| CN107563503A (en) | A kind of codified selects the design method that threshold values selects function artificial neuron | |
| CN107480783A (en) | A kind of more threshold values list functions select the design method of end output artificial neuron | |
| CN107480780A (en) | A kind of more threshold values polygamma functions select the design method of output artificial neuron more | |
| CN106682651A (en) | Monitoring method for cable insulation leather damage based on neural network | |
| CN107563504A (en) | A kind of double design methods for selecting end output artificial neuron of implantable activation primitive | |
| CN107506827A (en) | A kind of fixed neuronotropic design method of more threshold values polygamma functions | |
| CN107609636A (en) | A kind of polygamma function correspondingly exports the design method of feedback function artificial neuron | |
| CN107590536A (en) | A kind of graded potential formula selects the design method of end feedback artificial neuron | |
| CN107609640A (en) | A kind of threshold values selects the design method of end graded potential formula artificial neuron | |
| CN107480779A (en) | Design method that is a kind of while exporting polymorphic function artificial neuron |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180109 |