CN105786482A - Artificial intelligence system - Google Patents
Artificial intelligence system Download PDFInfo
- Publication number
- CN105786482A CN105786482A CN201610007922.4A CN201610007922A CN105786482A CN 105786482 A CN105786482 A CN 105786482A CN 201610007922 A CN201610007922 A CN 201610007922A CN 105786482 A CN105786482 A CN 105786482A
- Authority
- CN
- China
- Prior art keywords
- unit
- artificial intelligence
- improvement
- intelligence system
- administrative
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the field of artificial intelligence, and discloses an artificial intelligence system. The artificial intelligence system at least comprises a Manager, an Executor, and an Improver, wherein the Manager is connected with the Executor and the Improver, and is used to collect performance information of the artificial intelligence system; the Improver is used to generate an improvement plan according to the performance information collected by the Manager; the Manager is used to assign the improvement plan generated by the Improver to the Executor; the Executor is used for execution according to the improvement plan assigned by the Manager, so that the performance of the artificial intelligence system can be improved. The artificial intelligence system provided by the invention has the advantages that the Improver is used for performance information integration and autonomous formulation of the improvement plan, and the Executor is used for execution of the improvement plan, so that the potentials of hardware and software can be exploited, the system performance can be improved, and the artificial intelligence system can autonomously solve problems, which not only achieves strong artificial intelligence but also gives full play to the advantages of the conventional data processing technology.
Description
Technical field
The present invention relates to artificial intelligence field, particularly to a kind of artificial intelligence system.
Background technology
Artificial intelligence is intended to produce a kind of intelligent machine can made a response in the way of human intelligence is similar, to replace the mankind to think deeply, and high efficiency, low energy consumption and solve various problem intelligently, become a popular research topic at present.
Artificial intelligence is divided into weak artificial intelligence and the big class of strong artificial intelligence two, although input information can be made the reaction seeming intelligence by weak artificial intelligence, but does not have reasoning and autonomous problem-solving ability, it is impossible to realize real intellectuality.And artificial intelligence focuses on the brain of simulating human by force, can possess reasoning and autonomous problem-solving ability by replicating the activity of human brain, it is achieved intelligent.
In the process realizing the present invention, inventor have found that prior art at least exists following defect:
When strong artificial intelligence adopting purely the mode of simulation human brain realize intellectuality, the requirement of software and hardware is very high, and existing data processing technique can not be adopted, it is impossible to give full play to the advantage of available data treatment technology.
Summary of the invention
In order to solve problem of the prior art, embodiments provide a kind of artificial intelligence system.Described technical scheme is as follows:
First aspect, it is provided that a kind of artificial intelligence system, described artificial intelligence system includes:
Described artificial intelligence system at least includes: administrative unit, performance element and improvement unit, and described administrative unit is connected with described performance element, and described administrative unit is connected with described improvement unit；
Described administrative unit, for collecting the performance information of described artificial intelligence system；
Described improvement unit, for the performance information collected according to described administrative unit, generates improvement project；
Described administrative unit, for distributing to described performance element by the improvement project that described improvement unit generates；
Described performance element, performs for the improvement project distributed according to described administrative unit, to improve the performance of described artificial intelligence system.
In conjunction with first aspect, in the first possible implementation of first aspect, described artificial intelligence system also includes creating unit, and described creation unit is connected with described improvement unit；
Described improvement unit, for when generating improvement project failure based on existing instrument, sending to described creation unit and create instruction；
Described creation unit, for the creation instruction sent according to described improvement unit, creates new instrument, returns to described improvement unit；
Described improvement unit, for the new tool created based on described creation unit, generates improvement project.
The first in conjunction with first aspect is likely to implementation, and the second in first aspect is likely in implementation, and described artificial intelligence system also includes speech unit, and described speech unit is connected with described improvement unit；
Described improvement unit, when being additionally operable to generate improvement project failure if based on the new tool that described creation unit is created, by described speech unit, externally provides the hardware device list needed.
In conjunction with first aspect, in the third possible implementation of first aspect, described improvement unit is additionally operable to, when described artificial intelligence system is in idle condition, according to the performance information that described administrative unit is collected, generate described improvement project.
In conjunction with first aspect, in the 4th kind of possible implementation of first aspect, described artificial intelligence system also includes speech unit, and described speech unit is connected with described improvement unit；
Described speech unit, is used for receiving improvement instruction, and issues described improvement instruction to described improvement unit；
Described improvement unit, for the performance information collected according to described improvement instruction and described administrative unit, generates described improvement project.
In conjunction with the 4th kind of possible implementation of first aspect, in the 5th kind of possible implementation of first aspect, described speech unit is connected with described administrative unit；
Described speech unit, is additionally operable to receive work order, and issues described work order to described administrative unit；
Described administrative unit, is additionally operable to the work order by described speech unit issues and distributes to described performance element；
Described performance element, is additionally operable to perform described administrative unit assignment instruction.
In conjunction with any of the above-described kind of possible implementation of first aspect, in the 6th kind of possible implementation of first aspect, at least one unit in described artificial intelligence system has self-similar structure.
In conjunction with the 6th kind of possible implementation of first aspect, in the 7th kind of possible implementation of first aspect, described artificial intelligence system includes speech unit, and described speech unit is used for determining when any cell occurs abnormal, disconnects the described unit that exception occurs.
The technical scheme that the embodiment of the present invention provides has the benefit that
The artificial intelligence system that the embodiment of the present invention provides, by improving unit integration performance information, autonomous formulation improvement plan also performs improvement project by performance element, software and hardware potentiality can be excavated, improve systematic function, make this artificial intelligence system possess autonomous problem-solving ability, strong artificial intelligence can either be realized, it is also possible to give full play to the advantage of available data treatment technology.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme in the embodiment of the present invention, below the accompanying drawing used required during embodiment is described is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the premise not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the structural representation of a kind of artificial intelligence system that the embodiment of the present invention provides；
Fig. 2 is the structural representation of a kind of artificial intelligence system that the embodiment of the present invention provides
Fig. 3 is the structural representation of a kind of artificial intelligence system that the embodiment of the present invention provides；
Fig. 4 is the structural representation of a kind of artificial intelligence system that the embodiment of the present invention provides；
Fig. 5 is the structural representation of a kind of artificial intelligence system that the embodiment of the present invention provides；
Fig. 6 is the self-similar structure schematic diagram of a kind of artificial intelligence system that the embodiment of the present invention provides；
What Fig. 7 was that the embodiment of the present invention provides a kind of disconnects the structural representation of artificial intelligence system after anomaly unit；
Fig. 8 is a kind of reconfiguration of cell schematic diagram that the embodiment of the present invention provides.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under not making creative work premise, broadly fall into the scope of protection of the invention.
Before the embodiment of the present invention is described in detail, first the multiple unit that the present embodiments relate to are carried out description below explanation:
1, speech unit (Spokesperson is called for short S cell): externally, speech unit represents whole artificial intelligence system and carries out information exchange with the external world, it is possible to receives the instruction of external world's input and externally sends feedback；Internally, speech unit can interpolate that the instruction of extraneous input is work class instruction or improves class instruction, and the instruction that the external world inputs passes to administrative unit or improves unit；
2, administrative unit (Manager is called for short M unit): be responsible for distributing to work order corresponding performance element, and collect various performance informations for improving unit, perform to improve the improvement project of unit；
3, performance element (Executor, it is called for short E unit): only responsible receiving management unit assignment instruction, perform corresponding task, complete task or run in the process performing task after extremely, can feed back to administrative unit, it is also possible to submit to administrative unit to relevant system performance information；
4, improve unit (Improver is called for short I unit): unit creative in this artificial intelligence system, be responsible for finding and improve systematic function and adjust the artificial intelligence system method to complete Given task.Specifically include: connect existing unit in artificial intelligence system by new regulation, give performance element to use the new tool creating unit exploitation, create new unit etc.；
5, unit (Creator is called for short C cell) is created: be responsible for according to the Command design new tool improving unit.
Fig. 1 is the structural representation of a kind of artificial intelligence system that the embodiment of the present invention provides.Referring to Fig. 1, this artificial intelligence system includes: administrative unit 101, performance element 102 and improvement unit 103.This administrative unit 101 is connected with this performance element 102, and this administrative unit 101 is connected with this improvement unit 103.
This administrative unit 101, for collecting the performance information of this artificial intelligence system；
This improvement unit 103, for the performance information collected according to this administrative unit, generates improvement project；
This administrative unit 101, for distributing to this performance element by the improvement project that this improvement unit generates；
This performance element 102, performs for the improvement project distributed according to this administrative unit, to improve the performance of this artificial intelligence system.
In the implementation that the first is possible, this artificial intelligence system also includes creating unit, and this creation unit is connected with this improvement unit 103；
This improvement unit 103, for when generating improvement project failure based on existing instrument, sending to this creation unit and create instruction；
This creation unit, for the creation instruction sent according to this improvement unit 103, creates new instrument, returns to this improvement unit 103；
This improvement unit 103, for the new tool created based on this creation unit, generates improvement project.
In the implementation that the second is possible, this artificial intelligence system also includes speech unit, and this speech unit is connected with this improvement unit 103；
This improvement unit 103, when being additionally operable to generate improvement project failure if based on the new tool that this creation unit is created, by this speech unit, externally provides the hardware device list needed.
In the implementation that the third is possible, this improvement unit 103 is additionally operable to when this artificial intelligence system is in idle condition, according to the performance information that this administrative unit 101 is collected, generates this improvement project.
In the 4th kind of possible implementation, this artificial intelligence system also includes speech unit, and this speech unit is connected with this improvement unit 103；
This speech unit, is used for receiving improvement instruction, and issues this improvement instruction to this improvement unit 103；
This improvement unit 103, for the performance information collected according to this improvement instruction and this administrative unit 101, generates this improvement project.
In the 5th kind of possible implementation, this speech unit is connected with this administrative unit 101；
This speech unit, is additionally operable to receive work order, and issues this work order to this administrative unit 101；
This administrative unit 101, this performance element 102 is distributed in the work order being additionally operable to be issued by this speech unit；
This performance element 102, is additionally operable to perform this administrative unit 101 assignment instruction.
In the 6th kind of possible implementation, at least one unit in this artificial intelligence system has self-similar structure.
In the 7th kind of possible implementation, this artificial intelligence system includes speech unit, and this speech unit is used for determining when any cell occurs abnormal, disconnects the unit that this generation is abnormal.
Above-mentioned all optional technical schemes, it is possible to adopting and arbitrarily combine formation optional embodiment of the present invention, this is no longer going to repeat them.
Fig. 2 is the structural representation of a kind of artificial intelligence system that the embodiment of the present invention provides.Referring to Fig. 2, this artificial intelligence system at least includes: administrative unit 201, performance element 202 and improvement unit 203.This administrative unit 201 is connected with this performance element 202, and this administrative unit 201 is connected with this improvement unit 203.
Specifically, this artificial intelligence system can include one or more administrative unit 201, one or more performance element 202, one or more improvement unit 203, each unit therein can be made up of some computer software and hardware, and these unit carry out limited and controlled contact under the restriction of series of rules.
And an administrative unit 201 can be connected with one or more performance elements 202, it is also possible to be connected with one or more improvement unit 203.It addition, performance element 202 can be connected with improving unit 203, it is also possible to be not connected to.This is not all limited by the present embodiment.
This administrative unit 201 is for collecting the performance information of this artificial intelligence system, this improvement unit 203 performance information for collecting according to this administrative unit 201, generate improvement project, this administrative unit 201 is for distributing to this performance element 202 by the improvement project that this improvement unit 203 generates, this performance element 202 is for performing according to the improvement project of this administrative unit 201 distribution, to improve the performance of this artificial intelligence system.
Wherein, this performance information is for representing the performance that this artificial intelligence system is current, current systematic function or the performance of product produced under the control of this artificial intelligence system can be included, such as, this systematic function can include the CPU (CentralProcessingUnit that this artificial intelligence system is current, central processing unit) utilization rate, the bandwidth taken, task processing delay etc., these properties of product can include the yield rate of product, qualification rate etc., and this performance information is not limited by the present embodiment.
In embodiments of the present invention, this administrative unit 201 shares out the work task for this performance element 202, this performance element 202 can perform this administrative unit 201 assignment task, the performance information of this artificial intelligence system can be added up in the process of implementation, and report this administrative unit 201, this administrative unit 201 can collect the performance information of this artificial intelligence system, and stores this performance information.
In subsequent process, this improvement unit 203 can ask performance information to this administrative unit 201, and this performance information is analyzed, the performance that this artificial intelligence system is current is evaluated, according to analyzing result, attempt formulating the improvement project improving systematic function, the aspect pointing out the needs improvement of this artificial intelligence system and can improve, and this improvement project is sent to this administrative unit 201, by administrative unit 201, this improvement project is distributed to performance element 202, performance element 202 can perform according to this improvement project, thus improving the performance of this artificial intelligence system.
Wherein, the storage of this improvement unit 203 has series of algorithms, it is possible to adopt these algorithms to generate improvement project, and the algorithm adopted is not limited by the embodiment of the present invention.
And, this improvement unit 203 is when generating improvement project, improved model can be set up, attempt when without new tool and new hardware device, only by several performance elements 203 existing in reconnect system, generate improvement project by " the transferring " of performance element 203, reach the purpose improved.If generating improvement project success, this improvement project can be transferred to performance element 203 perform.
Alternatively, this improvement unit 203 is specifically for when this artificial intelligence system is in idle condition, according to the performance information that this administrative unit 201 is collected, generating this improvement project.
That is to say, this improvement unit 203 can analyze the running status that this artificial intelligence system is current, judge to be presently at busy state or idle condition, when determining that this artificial intelligence system is currently at idle condition, represent that performance element 202 is currently without performing task, then this improvement unit 203 can generate this improvement project according to this performance information, in order to performance element 202 performs according to improvement project.And when determining that this artificial intelligence system is currently at busy state, represent that performance element 202 currently performs task, then wouldn't generate improvement project.It is, of course, also possible to analyzed, according to the situation that this performance element 202 performs task, the running status that artificial intelligence system is current by this administrative unit 201, and will determine that result is sent to this improvement unit 203, this is not limited by the present embodiment.
The present embodiment illustrates only for administrative unit 201, performance element 202 and improvement unit 203, it practice, this artificial intelligence system could be included for realizing the unit of other functions, this is not limited by the present embodiment.
The artificial intelligence system that the embodiment of the present invention provides, the performance information of artificial intelligence system is integrated by improving unit, autonomous formulation improvement plan also performs improvement project by performance element, software and hardware potentiality can be excavated, improve systematic function, this artificial intelligence system is made to have possessed autonomous problem-solving ability, it is achieved that strong artificial intelligence.And, although this artificial intelligence system has redefined content and the compound mode of the information interaction of classical software and hardware, but the execution pattern of each component internal does not change, thus the available data treatment technology for single software and hardware can fully be applied in this artificial intelligence system insusceptibly, namely this artificial intelligence system can either realize strong artificial intelligence, it is also possible to gives full play to the advantage of available data treatment technology.
In the implementation that the first is possible, referring to Fig. 3, also including creating unit 204 in artificial intelligence system, this creation unit 204 is connected with this improvement unit 203, for creating new instrument.
This improvement unit 203 is for when generating improvement project failure based on existing instrument, sending to this creation unit 204 and create instruction；This creation unit 204, for the creation instruction sent according to this improvement unit 203, is created new instrument, is returned to this improvement unit 203；This improvement unit 203 new tool for creating based on this creation unit 204, generates improvement project.
nullSpecifically，The execution of improvement project needs to utilize the existing instrument of this artificial intelligence system，Therefore this improvement unit 203 can generate based on existing instrument when generating improvement project，When this improvement unit 203 generates improvement project failure based on existing instrument，Represent and cannot realize the improvement to this artificial intelligence system based on current existing instrument，Need new instrument，Now，This improvement unit 203 can send to this creation unit 204 and create instruction，This creation unit 204 is indicated to create new tool，This creation unit 204 can create new instrument when receiving this creation instruction，The new tool of creation is returned to this improvement unit 203，New tool can be added improved model by this improvement unit 203，Regenerate improvement project，Performance element 202 is transferred to perform again when generating improvement project success.
Wherein, can including document data base in this creation unit 204, this document data base is used for storing exploitation software program, and this creation unit 204 can pass through the exploitation software program stored, and creates new instrument.It addition, the instrument created in this artificial intelligence system can include printer interface, display interface etc., this is not limited by the present embodiment.
The artificial intelligence system that the embodiment of the present invention provides, new tool is independently created by creating unit, improvement project is generated based on new tool by improving unit, improve unit and creation unit jointly achieves various independence maintenance and is proactively improving the behavior of performance, this artificial intelligence system is made to have the characteristic realizing strong artificial intelligence, it is possible to progressively to realize performance boost and various intelligent characteristic in autonomous improvement.
In the implementation that the second is possible, referring to Fig. 4, this artificial intelligence system also includes speech unit 205, and this speech unit 205 interacts for the external world with this artificial intelligence system, it is possible to as the medium between internal element and the external world.
This speech unit 205 is connected with improving unit 203, when this improvement unit 203 is additionally operable to generate improvement project failure if based on the new tool that this creation unit 204 is created, is externally provided the hardware device list of needs by this speech unit 205.
Specifically, after creation unit 204 creates new tool, this improvement unit 203 can generate improvement project based on new tool, but when this improvement unit 203 generates improvement project failure based on new tool, even if representing that generating new tool can not realize the improvement to this artificial intelligence system, development receives the restriction of Current hardware equipment, now, this improvement unit 203 may determine that the hardware device list of needs, the hardware device list of needs is externally provided by this speech unit 205, so that attendant is according to this hardware device list, hardware device is added for this artificial intelligence system.Afterwards, this improvement unit 203 can generate improvement project based on the hardware device added.
Such as, this improvement unit 203 needs to improve the printing function of artificial intelligence system, printing interface is generated by creating unit 204, but owing to this artificial intelligence system does not configure printer, when generating improvement project based on printing interface, can still result in failed regeneration, now this improvement unit 203 can pass through speech unit 205 externally provides hardware device list, this hardware device list includes printer, when attendant adds printer for this artificial intelligence system, this improvement unit 203 can generate improvement project based on the printer of the printing interface generated and interpolation, thus with the addition of printing function for this artificial intelligence system.
Alternatively, after this improvement unit 203 can pass through the agreement that this speech unit 205 obtains extraneous attendant, then artificial intelligence system is improved.
That is to say, performance information can be analyzed by this improvement unit 203 when artificial intelligence system is currently at idle condition, attempt formulating the preliminary improvement project improving systematic function, extraneous attendant is submitted to by unit 205 of making a speech, if attendant agrees to this preliminary improvement project, this improvement unit 203 can be worked in coordination with creation unit 204 according to preliminary improvement project and be formulated formal improvement project, transfers to performance element 202 to perform.
Alternatively, this improvement unit 203 both can generate improvement project when artificial intelligence system is currently at idle condition, it is also possible to generates improvement project according to the order of speech unit 205.
Specifically, this speech unit 205 can receive the improvement instruction of extraneous input, and issue this improvement instruction to this improvement unit 203, when this improvement unit 203 receives this improvement instruction, the performance information that this administrative unit 201 is collected can be obtained, according to this improvement instruction and performance information, generate improvement project, this improvement project is sent to this administrative unit 201, by administrative unit 201, this improvement project is distributed to performance element 202, performance element 202 can perform according to this improvement project, thus improving the performance of this artificial intelligence system, meet the demand of attendant.
Alternatively, referring to Fig. 5, this speech unit 205 can also be connected with this administrative unit 201, issues work order to this administrative unit 201.
Specifically, this speech unit 205 can receive the work order of extraneous input, and issues work order to this administrative unit 201, and this work order can be distributed to performance element 202 by this administrative unit 201, performance element 202 can perform according to this work order, completes task.
In the process, this administrative unit 201 can follow the tracks of the Task Progress of performance element 202, collects relevant system performance information and exception record, and this performance element 202 is responsible for being embodied as, and reports the progress of task when needed to this administrative unit 201.When task completes, it is possible to reported to speech unit 205 by this administrative unit 201, speech unit 205 is responsible for making feedback to external world.When task occurs abnormal, when this performance element 202 decision instruction cannot perform, as required, speech unit 205 has two kinds of processing modes: external feedback task abnormity is not reprocessed this task, or automatically generate improvement task, attempt autonomous systematic function of improving to complete task.In task process, the external world can interrogate Task Progress at any time.
During practical application, this speech unit 205 can receive the instruction of extraneous input, judge that this instruction is that work order still improves instruction, if this instruction is work order, distribute to this administrative unit 201, if this instruction is for improving instruction, distribute to this improvement unit 203.
The artificial intelligence system that the embodiment of the present invention provides, can be realized basic externally mutual by unit of making a speech, and except spokesman's unit, all unit directly can not contact with the external world, this makes the input/output signal of artificial intelligence system can be easily and effectively monitored, substantially increase outer bound pair artificial intelligence system and implement the difficulty of malicious attack, the difficulty causing actual destruction when also making artificial intelligence system out of control to external world increases, these all make artificial intelligence system while having highly intelligent and complexity, still controlled and safety.
In the implementation that the third is possible, at least one unit in this artificial intelligence system has self-similar structure.
Represent speech unit with S, represent administrative unit with M, represent performance element with E, represent improvement unit with I, represent creation unit with C.Then this artificial intelligence system has ESMIC framework, it is possible to be called ESMIC system.Referring to Fig. 6, can including E, S, M, I, C cell in this artificial intelligence system, M unit therein can also include ME, MS, MM, MI, mc unit, and namely M unit has self-similar structure.
When a certain unit in this artificial intelligence system has self-similar structure, this unit itself has strong artificial intelligence, it is possible to achieve independently improve performance.
Especially when improving unit and there is self-similar structure, improve unit itself and there is strong artificial intelligence, can formulate more suitably improvement project, and the potential demand improved from feedback opinion learning the prognoses system of attendant and direction, thus being greatly improved the performance of artificial intelligence system.Owing to having spontaneous improvement ability, the performance boost of system even can early than the artificial proposition improving instruction, the ESMIC system in other words with self-similar structure can learn the changes in demand of prediction user, provide for a rainy day, the inefficient situation reducing " embracing Buddha's feet in one's hour of need " occurs, and the characteristic remarkable of this height intelligence improves the overall performance of system.
Five unit in this artificial intelligence system all can possess self-similar structure, such it is advantageous in that: can cooperate between unit, even if the performance of a certain unit is undesirable, overall performance can also be ensured by the intellectuality of other unit, greatly strengthen antagonism software and hardware and attack the resistance with system exception, improve system reliability and process performance so that this artificial intelligence system has splendid extension and self organization ability.
Further, the software and hardware in this artificial intelligence system is under attack or when damaging, some unit can occur feedback abnormal, and for safeguards system safety, this artificial intelligence system must actively disconnect the data cube computation with anomaly unit as early as possible, reattempts repair afterwards.The artificial intelligence system now with self-similar structure will play its advantage:
If the first module in this artificial intelligence system occurs abnormal, it is abnormal that first module or other unit being connected with first module can detect that this first module occurs, and abnormal conditions are reported this speech unit 205, now this speech unit 205 determines that this first module occurs extremely, it is possible to disconnect the connection of this first module and other unit.
Now, if the second unit except this first module has self-similar structure, include the unit identical with first module function in this second unit, then in this second unit, the unit identical with first module function can replace first module, completes the function of this first module.
As shown in Figure 6, if I, E and C cell are all destroyed or exception occurs, I, E and C cell are disconnected, as shown in Figure 7.M unit owing to having ESMIC structure is not destroyed, S cell by the S cell " combination " with M unit subordinate, jointly completes the task of S cell after noting abnormalities, and other parts in M unit will substitute the appropriate section in original system, realize its function, as shown in Figure 8.
The artificial intelligence system that the embodiment of the present invention provides, by dynamic reorganization unit connection structure with independently write the task of cannot completing before new tool completes, excavates software and hardware potentiality, promotes performance during execution similar tasks next time.This makes be provided with entirely autonomous solution on system parenchyma and propose the ability of problem, it is achieved that strong artificial intelligence.When a part for system is destroyed or disconnects, as long as also remaining has complete ESMIC subsystem, this system would not be collapsed, and can keep characteristic and the foreign exchanges function of ESMIC system.
Additionally, several systems loading ESMIC framework can independently be combined into a new system easily and complete task, new system has the data of original system and function simultaneously and is optimized, and the service that each system external provides in the process of combination is without interruption.
In the embodiment of the present invention, in the incipient stage that this artificial intelligence system is implemented, being above-mentioned five functional units by hardware-software partition, each write realizes program and the interaction mechanism of its basic function, and completes to connect.This artificial intelligence system can realize strong artificial intelligence by above-mentioned five functional units, constantly this artificial intelligence system is improved.
Improvement unit therein and creation unit the execution being not involved in task, can remove at any time from this artificial intelligence system.When this artificial intelligence system has been improved, when needing to carry out volume production, in order to ensure safety, the improvement unit in this artificial intelligence system can be removed and create unit, only retain speech unit, administrative unit and performance element, now this artificial intelligence system no longer has improvement ability, that can show and traditional computer seemingly, its artificial intelligence's level is also by decline and be maintained at weak artificial intelligence's level.
Additionally, the improvement unit improved in unit with self similarity framework can also be removed and create unit, only retain and there is the speech unit improved in unit of self similarity framework, administrative unit and performance element, and fix the inspection last inspection together as various improvement plans for it, now system will have controlled improvement ability, not havinging a structural change with menace and situation that new tool transfers to executor to perform, improving and safety depositing thus realizing system autonomous.Wherein, the fixing constraints checked can be robot three big law or other constraints etc..
In sum, the artificial intelligence system that the embodiment of the present invention provides, by dynamic reorganization unit connection structure with independently write cannot completing before new tool completes ESMIC system of task, the performance data passed through before integrating independently can formulate improvement plan and perform, excavate software and hardware potentiality, promote performance when performing similar tasks next time.This new software and hardware architecture be applicable to various at present can on interconnective software and hardware process equipment so that it is there is entirely autonomous solution and propose the ability of problem, be i.e. strong artificial intelligence.And, although artificial intelligence system has redefined content and the compound mode of the information interaction of classical software and hardware, but the execution pattern of each component internal does not change, thus the available data treatment technology for single software and hardware can fully be applied in this artificial intelligence system insusceptibly.
Further, all unit except spokesman's unit directly do not contact with the external world, this makes the input/output signal of system can be easily and effectively monitored, which greatly enhances outer bound pair system and implement the difficulty of malicious attack, when also making intelligence system out of control, cause the difficulty of actual destruction to increase to external world.These all make ESMIC system while having highly intelligent and complexity, still controlled and safety.
Further, software system complete exploitation carry out volume production time, remove the improvement unit with improvement ability and create unit, to ensure that system has controlled improvement ability, do not have a structural change with menace and situation that new tool transfers to executor to perform, improving and safety depositing thus realizing system autonomous, can safely and effectively control its level of intelligence and behavior.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all within the spirit and principles in the present invention, any amendment of making, equivalent replacement, improvement etc., should be included within protection scope of the present invention.
Claims (8)
1. an artificial intelligence system, it is characterised in that described artificial intelligence system at least includes: administrative unit, performance element and improvement unit, described administrative unit is connected with described performance element, and described administrative unit is connected with described improvement unit；
Described administrative unit, for collecting the performance information of described artificial intelligence system；
Described improvement unit, for the performance information collected according to described administrative unit, generates improvement project；
Described administrative unit, for distributing to described performance element by the improvement project that described improvement unit generates；
Described performance element, performs for the improvement project distributed according to described administrative unit, to improve the performance of described artificial intelligence system.
2. artificial intelligence system according to claim 1, it is characterised in that described artificial intelligence system also includes creating unit, and described creation unit is connected with described improvement unit；
Described improvement unit, for when generating improvement project failure based on existing instrument, sending to described creation unit and create instruction；
Described creation unit, for the creation instruction sent according to described improvement unit, creates new instrument, returns to described improvement unit；
Described improvement unit, for the new tool created based on described creation unit, generates improvement project.
3. artificial intelligence system according to claim 2, it is characterised in that described artificial intelligence system also includes speech unit, and described speech unit is connected with described improvement unit；
Described improvement unit, when being additionally operable to generate improvement project failure if based on the new tool that described creation unit is created, by described speech unit, externally provides the hardware device list needed.
4. artificial intelligence system according to claim 1, it is characterised in that described improvement unit is additionally operable to, when described artificial intelligence system is in idle condition, according to the performance information that described administrative unit is collected, generate described improvement project.
5. artificial intelligence system according to claim 1, it is characterised in that described artificial intelligence system also includes speech unit, and described speech unit is connected with described improvement unit；
Described speech unit, is used for receiving improvement instruction, and issues described improvement instruction to described improvement unit；
Described improvement unit, for the performance information collected according to described improvement instruction and described administrative unit, generates described improvement project.
6. artificial intelligence system according to claim 5, it is characterised in that described speech unit is connected with described administrative unit；
Described speech unit, is additionally operable to receive work order, and issues described work order to described administrative unit；
Described administrative unit, is additionally operable to the work order by described speech unit issues and distributes to described performance element；
Described performance element, is additionally operable to perform described administrative unit assignment instruction.
7. the artificial intelligence system according to any one of claim 1-6, it is characterised in that at least one unit in described artificial intelligence system has self-similar structure.
8. artificial intelligence system according to claim 7, it is characterised in that described artificial intelligence system includes speech unit, described speech unit is used for determining when any cell occurs abnormal, disconnects the described unit that exception occurs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610007922.4A CN105786482A (en) | 2016-01-06 | 2016-01-06 | Artificial intelligence system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610007922.4A CN105786482A (en) | 2016-01-06 | 2016-01-06 | Artificial intelligence system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105786482A true CN105786482A (en) | 2016-07-20 |
Family
ID=56390417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610007922.4A Pending CN105786482A (en) | 2016-01-06 | 2016-01-06 | Artificial intelligence system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105786482A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108508826A (en) * | 2018-03-20 | 2018-09-07 | 广州大学 | A kind of apparatus and system for preventing artificial intelligence behavior body out of control |
| WO2019037077A1 (en) * | 2017-08-25 | 2019-02-28 | 深圳市得道健康管理有限公司 | Method for establishing artificial intelligence-based behavior control database and device and system thereof |
| CN109506714A (en) * | 2018-12-22 | 2019-03-22 | 吴碧玉 | A kind of artificial intelligence bridge security detection system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102740326A (en) * | 2011-04-07 | 2012-10-17 | 中国移动通信集团公司 | Managing method and device for processing resources in BBU |
| US20140068627A1 (en) * | 2012-08-31 | 2014-03-06 | Silicon Graphics International Corp. | Dynamic resource scheduling |
| CN104239150A (en) * | 2014-09-15 | 2014-12-24 | 杭州华为数字技术有限公司 | Method and device for adjusting hardware resources |
| CN104932926A (en) * | 2015-07-09 | 2015-09-23 | 上海联彤网络通讯技术有限公司 | System and method for achieving dynamic matching of equipment and drives in intelligent operation system platform |
-
2016
- 2016-01-06 CN CN201610007922.4A patent/CN105786482A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102740326A (en) * | 2011-04-07 | 2012-10-17 | 中国移动通信集团公司 | Managing method and device for processing resources in BBU |
| US20140068627A1 (en) * | 2012-08-31 | 2014-03-06 | Silicon Graphics International Corp. | Dynamic resource scheduling |
| CN104239150A (en) * | 2014-09-15 | 2014-12-24 | 杭州华为数字技术有限公司 | Method and device for adjusting hardware resources |
| CN104932926A (en) * | 2015-07-09 | 2015-09-23 | 上海联彤网络通讯技术有限公司 | System and method for achieving dynamic matching of equipment and drives in intelligent operation system platform |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019037077A1 (en) * | 2017-08-25 | 2019-02-28 | 深圳市得道健康管理有限公司 | Method for establishing artificial intelligence-based behavior control database and device and system thereof |
| CN108508826A (en) * | 2018-03-20 | 2018-09-07 | 广州大学 | A kind of apparatus and system for preventing artificial intelligence behavior body out of control |
| CN109506714A (en) * | 2018-12-22 | 2019-03-22 | 吴碧玉 | A kind of artificial intelligence bridge security detection system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112668237B (en) | Digital twin model based on industrial Internet cloud platform and construction method thereof | |
| Gomaa et al. | Software adaptation patterns for service-oriented architectures | |
| Dai et al. | Reliability analysis of grid computing systems | |
| CN114048857B (en) | Calculation force distribution method and device and calculation force server | |
| CN103581339A (en) | Storage resource allocation monitoring and processing method based on cloud computing | |
| Hu et al. | Adaptive slave controller assignment for fault-tolerant control plane in software-defined networking | |
| CN105786482A (en) | Artificial intelligence system | |
| CN107273589A (en) | Reconstruction strategy generation system and its generation method based on DIMA systems | |
| JP2010541100A (en) | How to implement highly available data parallel operations on a computational grid | |
| CN101621404B (en) | Method and system for layering processing of failure | |
| Kumar et al. | Optimal multilevel redundancy allocation in series and series–parallel systems | |
| CN113325813A (en) | Embedded multi-robot cooperative task allocation and control system and device | |
| CN117349014A (en) | Model training method and device based on cluster resources, electronic equipment and medium | |
| CN109344059B (en) | Server pressure testing method and device | |
| CN116578412A (en) | Machine learning cluster computing power resource operation and maintenance method, system, equipment and storage medium | |
| CN105471986A (en) | Data center construction scale assessment method and apparatus | |
| CN105550065A (en) | Database server communication management method and device | |
| Majstorović et al. | Industry 4.0 in Serbia-state of development | |
| CN204904019U (en) | PAS100 control system's overall redundant framework | |
| CN105450751B (en) | A kind of system improving cloud computing environment stability | |
| CN110391929A (en) | A kind of fault tolerant control method, device and fault-tolerant component | |
| CN115002217B (en) | Scheduling method, device, equipment and medium | |
| CN115460217B (en) | Cloud service high availability decision-making method based on reinforcement learning | |
| JP7391809B2 (en) | Apparatus, method and program for determining validity of network status confirmation procedure | |
| CN107276828A (en) | A kind of dispatching method and device of cluster interior joint |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160720 |
|
| RJ01 | Rejection of invention patent application after publication |