CN117057162B - Task execution method and device, storage medium and electronic equipment - Google Patents
Task execution method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN117057162B CN117057162B CN202311307471.2A CN202311307471A CN117057162B CN 117057162 B CN117057162 B CN 117057162B CN 202311307471 A CN202311307471 A CN 202311307471A CN 117057162 B CN117057162 B CN 117057162B
- Authority
- CN
- China
- Prior art keywords
- data
- supplementary
- distribution
- task
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000003860 storage Methods 0.000 title claims abstract description 22
- 238000009826 distribution Methods 0.000 claims abstract description 137
- 238000005070 sampling Methods 0.000 claims abstract description 39
- 230000006870 function Effects 0.000 claims description 43
- 238000004590 computer program Methods 0.000 claims description 16
- 230000001502 supplementing effect Effects 0.000 claims description 15
- 230000000153 supplemental effect Effects 0.000 claims description 10
- 238000005457 optimization Methods 0.000 claims description 4
- 230000003993 interaction Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 12
- 230000006872 improvement Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000013178 mathematical model Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000010979 ruby Substances 0.000 description 1
- 229910001750 ruby Inorganic materials 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Complex Calculations (AREA)
Abstract
The specification discloses a task execution method, a task execution device, a storage medium and electronic equipment. The task execution method comprises the following steps: the method comprises the steps of obtaining a target task model and expert experience data and historical task data corresponding to the target task model, determining the distribution of values of each dimension of the historical task data in the value range of input parameters of the target task model, taking the distribution as the value distribution, sampling from the value range of the input parameters according to the expert experience data and the value distribution to obtain all the supplementary input parameters, inputting all the supplementary input parameters into the target task model, obtaining all the supplementary output parameters corresponding to all the supplementary input parameters, constructing decision suggestion distribution according to all the supplementary input parameters, all the supplementary output parameters and the historical task data, determining optimal input parameters according to the decision suggestion distribution, and executing tasks according to the optimal input parameters.
Description
Technical Field
The present disclosure relates to the field of data analysis technologies, and in particular, to a task execution method, a task execution device, a storage medium, and an electronic device.
Background
With the development of computer technology, methods for constructing mathematical models for performing task execution by analyzing historical data are widely used in various fields.
In general, when a mathematical model is constructed, a large amount of historical task data of a designated task needs to be obtained as sample data, and the constructed model is trained and optimized based on the sample data, however, there are few historical task data which can be obtained in a part of designated tasks, and the cost for obtaining the historical task data is high, so that the accuracy of the constructed model is low.
Disclosure of Invention
The present disclosure provides a task execution method, a task execution device, a storage medium, and an electronic device, so as to partially solve the foregoing problems in the prior art.
The technical scheme adopted in the specification is as follows:
the specification provides a task execution method, which comprises the following steps:
acquiring a target task model, expert experience data and historical task data corresponding to the target task model, wherein the expert experience data is used for representing the confidence level of data of each dimension of input parameters of the target task model on the output result obtained by the target task model; determining the distribution of the value of each dimension of the historical task data in the value range of the input parameter of the target task model as a value distribution;
Sampling from the value range of the input parameters according to the expert experience data and the value distribution to obtain each supplementary input parameter; inputting the supplementary input parameters into the target task model to obtain supplementary output parameters corresponding to the supplementary input parameters;
and constructing decision suggestion distribution according to the supplementary input parameters, the supplementary output parameters and the historical task data, determining optimal input parameters according to the decision suggestion distribution, and executing tasks according to the optimal input parameters.
Optionally, acquiring the target task model and the historical task data specifically includes:
acquiring an initial task model and historical task data;
acquiring error distribution of the output value of the initial task model as prior error distribution;
according to the error between the output value obtained by inputting the historical task data into the initial task model and the actual output value, optimizing the prior error distribution to obtain posterior error distribution;
and optimizing the initial task model according to the posterior error distribution to obtain a target task model.
Optionally, according to the expert experience data and the value distribution, sampling from the value range of the input parameters to obtain each supplementary input parameter, which specifically includes:
Determining the distribution of the weight of each dimension of the input parameters of the target task model according to the expert experience data, wherein the weight of each dimension is used for representing the importance degree of the data of the dimension to the output result of the target task model as an importance distribution;
and sampling from the value range of the input parameters according to the importance distribution and the value distribution to obtain each supplementary input parameter.
Optionally, according to the expert experience data and the value distribution, sampling from the value range of the input parameters to obtain each supplementary input parameter, which specifically includes:
determining a sampling probability density function corresponding to the value range of the input parameter according to the expert experience data and the value distribution, wherein the sampling probability density function is used for representing the probability that each value in the value range of each dimension of the input parameter is sampled;
and sampling from the value range of the input parameters according to the sampling probability density function to obtain each supplementary input parameter.
Optionally, constructing a decision suggestion distribution according to the supplementary input parameters, the supplementary output parameters and the historical task data, which specifically includes:
Taking the supplementary input parameters and the supplementary output parameters corresponding to the supplementary data parameters as supplementary task data;
sending the supplementary task data to a designated device, displaying the supplementary task data to a designated user through the designated device, and enabling the designated user to determine a confidence level corresponding to the supplementary task data for each supplementary task data, wherein the confidence level is used for characterizing the accuracy of the supplementary output parameters in the supplementary task data obtained by the target task model according to the supplementary input parameters in the supplementary task data;
and constructing decision suggestion distribution according to the confidence coefficient and the historical task data.
Optionally, constructing a decision suggestion distribution according to the confidence level and the historical task data, which specifically includes:
constructing confidence distribution according to the confidence coefficient corresponding to each supplementary task data;
constructing decision weight distribution of each supplementary task data according to the confidence coefficient distribution and the expert experience data, wherein the decision weight distribution is used for representing the importance degree of each supplementary task data for optimizing the target task model;
And constructing decision suggestion distribution according to the decision weight distribution and the historical task data.
Optionally, determining an optimal input parameter according to the decision suggestion distribution, and performing task execution according to the optimal input parameter, which specifically includes:
sampling from the value range of the input parameters based on the decision suggestion distribution to obtain each sample input parameter and the probability that each sample input parameter is the optimal input parameter;
fitting the probability that the input parameter of each sample is the optimal input parameter to obtain an input probability density function;
and determining optimal input parameters according to the input probability density function, and executing tasks according to the optimal input parameters.
The present specification provides a task execution device including:
the acquisition module is used for acquiring a target task model, expert experience data and historical task data corresponding to the target task model, wherein the expert experience data is used for representing the confidence level of data of each dimension of input parameters of the target task model to the output result of the target task model;
a determining module, configured to determine a distribution of values of each dimension of the historical task data in a value range of an input parameter of the target task model as a value distribution;
The first supplementary module is used for sampling each supplementary input parameter from the value range of the input parameter according to the expert experience data and the value distribution;
the second supplementing module is used for inputting the supplementing input parameters into the target task model to obtain the supplementing output parameters corresponding to the supplementing input parameters;
and the execution module is used for constructing decision suggestion distribution according to the supplementary input parameters, the supplementary output parameters and the historical task data, determining optimal input parameters according to the decision suggestion distribution, and executing tasks according to the optimal input parameters.
The present specification provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the task execution method described above.
The present specification provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the task execution method described above when executing the program.
The above-mentioned at least one technical scheme that this specification adopted can reach following beneficial effect:
In the task execution method provided by the specification, firstly, a target task model and expert experience data and historical task data corresponding to the target task model are obtained, the expert experience data are used for representing the confidence level of data of each dimension of input parameters of the target task model to an output result of the target task model, the distribution of values of each dimension of the historical task data in a value range of the input parameters of the target task model is determined and used as the value distribution, each supplementary input parameter is obtained by sampling from the value range of the input parameters according to the expert experience data and the value distribution, each supplementary input parameter is input to the target task model, each supplementary output parameter corresponding to each supplementary input parameter is obtained, the target task model is optimized according to each supplementary input parameter and each supplementary output parameter, the optimized target task model is obtained, and task execution is carried out through the optimized target task model.
According to the method, expert experience data and historical task data can be combined to construct a model, so that accuracy of the model constructed under the condition that sample data are fewer can be effectively improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the specification, illustrate and explain the exemplary embodiments of the present specification and their description, are not intended to limit the specification unduly. In the drawings:
FIG. 1 is a schematic flow chart of a task execution method provided in the present specification;
FIG. 2 is a schematic illustration of an optimization process of the initial task model provided in this specification;
FIG. 3 is a schematic diagram of a decision suggestion distribution determination process provided in the present specification;
FIG. 4 is a schematic diagram of a task performing device provided in the present specification;
fig. 5 is a schematic diagram of an electronic device corresponding to fig. 1 provided in the present specification.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the present specification more apparent, the technical solutions of the present specification will be clearly and completely described below with reference to specific embodiments of the present specification and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present specification. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are intended to be within the scope of the present disclosure.
The following describes in detail the technical solutions provided by the embodiments of the present specification with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a task execution method provided in the present specification, including the following steps:
s101: and acquiring a target task model, expert experience data and historical task data corresponding to the target task model, wherein the expert experience data is used for representing the confidence level of data of each dimension of input parameters of the target task model on the output result obtained by the target task model.
In the specification, when a task is required to be executed, the service platform can firstly acquire basic data of a target task required to be executed, and construct an initial task model according to the acquired basic data, so that the target task model can be obtained by optimizing the initial task model, and further task execution can be performed through the target task model.
For example: assuming that the objective task is to control the robotic arm to perform an operation of resecting the diseased portion of the patient, the basic data described above include: the interaction relation data among the control instructions which can be input by the mechanical arm and the excision situation data of the diseased part of the patient corresponding to the control instructions which are input by the mechanical arm in a history way, wherein the control instructions which can be input are used for controlling each joint of the mechanical arm to be adjusted according to the appointed adjusting parameters, and the constructed task model can recommend a group of instructions for controlling the mechanical arm to perform operation for a doctor according to the excision situation data of the diseased part of the patient.
For another example: assuming that the target task is to generate a recommended marketing plan for a company, the basic data includes: historically, the effect data of the feasible marketing strategies of each commodity (such as strategies of time-limited price reduction, time-limited discount and buying two-sending and the like) and the influence relation data of other commodities when different commodities adopt the marketing strategies (such as that when one shirt adopts the time-limited discount, the sales of other shirts are influenced), and a user of a company can input sales targets of the quarter (such as that the sales of the quarter reaches the appointed amount and the like) into the constructed task model so as to determine the marketing strategy which can be adopted by each commodity for the company through the task model.
In the actual application scenario, the obtained basic data for constructing the initial task model is less, so that the accuracy of the constructed initial task model is lower, and therefore, the service platform can optimize the initial task model based on the obtained historical task data to obtain the target task model, as shown in fig. 2.
FIG. 2 is a schematic illustration of an optimization process for an initial task model provided in this specification.
As can be seen from fig. 2, the server may obtain historical task data, and obtain an error distribution of an output value of the initial task model as a priori error distribution, and optimize the priori error distribution according to an error between the output value obtained by inputting the historical task data into the initial task model and an actual output value to obtain a posterior error distribution, so as to optimize the initial task model according to the posterior error distribution to obtain a target task model, which may be specifically referred to by the following formula:
in the above-mentioned formula(s),to be used inHistorical task data is input to the resulting output values in the initial task model,is the actual output value (i.e. the true value),>an error value of the value is output for the initial task model.
Further, for the obtained error valueAssume that the error satisfies a priori of the gaussian distributionThe radial basis functions (Radial Basis Function, RBF) can be used as gaussian process kernels (kernel) to model the covariance between the dimensions of the historical task data. Wherein the gaussian process kernel can refer to the following formula:
according to the historical task dataA kernel matrix is established, and the following formula can be specifically referred to:
and then the inverse of the kernel matrix can be calculated by a square root method, gaussian process regression is performed, and the mean value and standard deviation equation of the error are obtained by deduction based on the prior error distribution.
In the above-mentioned formula(s),is the mean value of error, +.> As a variance of the error, the error is,representing the error value of the known data set.
Further, a posterior error distribution can be obtained, wherein the posterior error distribution can be referred to by the following formula:
it should be noted that the initial task model may be a neural network model or other mathematical models.
Further, the service platform can acquire the target task model, expert experience data and historical task data corresponding to the target task model, construct decision suggestion distribution, determine optimal input parameters according to the constructed decision suggestion distribution, and execute tasks according to the optimal input parameters, wherein the expert experience data are used for representing confidence levels of data of each dimension of the input parameters of the target task model on the output results of the target task model.
In the present specification, the execution body of the method for implementing task execution may refer to a designated device provided on a service platform by a server or the like, or may refer to a terminal device such as a desktop computer, a notebook computer, or the like, and for convenience of description, the method for implementing task execution provided in the present specification will be described below by taking the server as an example of the execution body.
S102: and determining the distribution of the value of each dimension of the historical task data in the value range of the input parameter of the target task model as a value distribution.
S103: and sampling from the value range of the input parameters according to the expert experience data and the value distribution to obtain each supplementary input parameter.
Further, the server may determine, as the value distribution, a distribution of the value of each dimension of the historical task data in the value range of the input parameter of the target task model through a preset exploration function, where the exploration function may refer to the following formula:
in the above-mentioned formula(s),for exploratory function, ++>Wherein, for each value in the value range of the input parameter of the target task model, if the value has corresponding historical task data, the value corresponds +.>The value of (2) is 1, if the value does not have the corresponding historical task data, the value corresponds to +.>The value of (2) is 0.
Further, the server may determine, according to expert experience data, a distribution of weights of each dimension of the input parameters of the target task model, as an importance distribution, for each dimension, the weights of the dimension being used to characterize a degree of importance of the data of the dimension to the output result obtained by the target task model.
Specifically, the server may determine, according to expert experience data and a value distribution, a sampling probability density function corresponding to a value range of the input parameter, and sample, according to the sampling probability density function, each supplementary input parameter from the value range of the input parameter, where the sampling probability density function is used to characterize a probability that each value in the value range of each dimension of the input parameter is sampled, and specifically may refer to the following formula:
in the above-mentioned formula(s),sampling probability density function corresponding to the value range of the input parameter,/, for>For value distribution->Is an importance distribution.
For example: assume that the value range of the input parameters is 0-100, wherein if the input parameters are distributed according to the valuesThe historical task data distributed in the range of 20-30 of the value range of the input parameter is determined to be less, and the input parameter is distributed according to the importance +.>And determining that the range of the value range of the input parameter is 20-30 has higher importance degree on the output result obtained by the target task model, and the probability that each value in the range of the value range of the input parameter is 20-30 is sampled is higher.
Further, the server may sample the supplemental input parameters from a range of values for the input parameters according to a sampling probability density function.
The server samples from the value range of the input parameter according to the sampling probability density function by using an acceptance-rejection sampling method based on the obtained sampling probability density function.
In particular, for values contained in the value range of the input dataTo->Whether the probability of the data is the sampled data point is judged, after each sampling to obtain a new data point, the value distribution is reconstructed, and the exploration data set is updatedAnd exploratory function->The accept-reject method is reused until m supplementary input parameters of the required specified data amount are obtained, and the supplementary input parameter sets are obtained>。
S104: and inputting the supplementary input parameters into the target task model to obtain supplementary output parameters corresponding to the supplementary input parameters.
Further, the server may input each of the supplemental input parameters to the target task model to obtain each of the supplemental output parameters corresponding to each of the supplemental input parameters.
S105: and constructing decision suggestion distribution according to the supplementary input parameters, the supplementary output parameters and the historical task data, determining optimal input parameters according to the decision suggestion distribution, and executing tasks according to the optimal input parameters.
In this specification, the server may send each supplementary input parameter and a supplementary output parameter corresponding to each supplementary data parameter as each supplementary task data, send each supplementary task data to the designated device, so as to display each supplementary task data to a designated user (for example, an administrator, a developer, etc.) through the designated device, and enable the designated user to determine, for each supplementary task data, a confidence level corresponding to the supplementary task data, and construct a decision suggestion distribution according to the determined confidence level and historical task data, where the confidence level is used to characterize the accuracy of the target task model to obtain the supplementary output parameters in the supplementary task data according to the supplementary input parameters in the supplementary task data, as shown in fig. 3.
Fig. 3 is a schematic diagram of a decision suggestion distribution determination process provided in the present specification.
As can be seen in conjunction with fig. 3, the server may construct a confidence distribution according to the confidence levels corresponding to the respective supplementary task data, where the confidence distribution may refer to the following formula:
further, the server may construct a decision weight distribution of each supplementary task data according to the confidence distribution and expert experience data, where the decision weight distribution is used to characterize the importance degree of each supplementary task data to the optimization target task model, and specifically may refer to the following formula:
in the above-mentioned formula(s),for decision weight distribution, ∈>For confidence distribution, ++>For the importance distribution determined from expert experience data.
Further, the server may construct a decision suggestion distribution according to the decision weight distribution and the historical task data, and the following formula may be specifically referred to:
in the above-mentioned formula(s),for decision weight distribution, ∈>，/>Is the target interval, i.e. the value range of the output parameter.
Further, the server may sample from a value range of input parameters based on a decision suggestion distribution to obtain each sample input parameter, and fit the probability of each sample input parameter being an optimal input parameter to obtain an input probability density function, and determine the input parameter according to the input probability density function, and perform task execution according to the optimal input parameter, where the input probability density function is used to characterize the probability of whether each value in the value range of the input parameter of the target task model is an optimal input parameter.
In the foregoing, the task execution performed by the server according to the optimal input parameters may be to send the determined optimal input parameters to the device used by the user, so as to display the optimal input parameters to the user through the device used by the user.
From the foregoing, it can be seen that the server can sample the input probability density function based on the decision suggestion distribution by the Markov chain Monte Carlo algorithm, and in particular, the server can randomly generate a new input parameter based on the value range of the input parameterCalculate->To measure whether the currently generated input parameter is accepted or rejected +.>As the probability of sampling a sample, a uniform distribution-compliant +.>Random number +.>If (if)Accept->And regenerate->For the next round of sampling.
If it isReject->And regenerate->And carrying out the next round of sampling until the end condition is reached, and fitting the sampling result by the method to obtain the input probability density function.
According to the method, the server can integrate expert experience data into the model based on a Bayesian statistical method to optimize the existing mathematical model, so that an input probability density function corresponding to each input parameter is obtained, and further, the optimal input parameter can be determined based on the input probability density function, and task execution can be performed based on the optimal input parameter.
The above is a method for implementing task execution for one or more embodiments of the present disclosure, and based on the same concept, the present disclosure further provides a corresponding task execution device, as shown in fig. 4.
Fig. 4 is a schematic diagram of a task execution device provided in the present specification, including:
the acquiring module 401 is configured to acquire a target task model, and expert experience data and historical task data corresponding to the target task model, where the expert experience data is used to characterize a confidence level of data of each dimension of an input parameter of the target task model to an output result obtained by the target task model;
a determining module 402, configured to determine a distribution of values of each dimension of the historical task data in a value range of an input parameter of the target task model as a value distribution;
a first supplementing module 403, configured to sample each supplementary input parameter from the value range of the input parameter according to the expert experience data and the value distribution;
a second supplementing module 404, configured to input the each supplementing input parameter to the target task model, so as to obtain each supplementing output parameter corresponding to the each supplementing input parameter;
and the execution module 405 is configured to construct a decision suggestion distribution according to the supplementary input parameters, the supplementary output parameters and the historical task data, determine an optimal input parameter according to the decision suggestion distribution, and execute a task according to the optimal input parameter.
Optionally, the acquiring module 401 is specifically configured to acquire an initial task model and historical task data; acquiring error distribution of the output value of the initial task model as prior error distribution; according to the error between the output value obtained by inputting the historical task data into the initial task model and the actual output value, optimizing the prior error distribution to obtain posterior error distribution; and optimizing the initial task model according to the posterior error distribution to obtain a target task model.
Optionally, the first supplementing module 403 is specifically configured to determine, according to the expert experience data, a distribution of weights of each dimension of the input parameters of the target task model, as an importance distribution, where, for each dimension, the weight of the dimension is used to characterize an importance degree of the data of the dimension to the output result obtained by the target task model; and sampling from the value range of the input parameters according to the importance distribution and the value distribution to obtain each supplementary input parameter.
Optionally, the first supplementing module 403 is specifically configured to determine, according to the expert experience data and the value distribution, a sampling probability density function corresponding to a value range of the input parameter, where the sampling probability density function is used to characterize a probability that each value in the value range of each dimension of the input parameter is sampled; and sampling from the value range of the input parameters according to the sampling probability density function to obtain each supplementary input parameter.
Optionally, the executing module 405 is specifically configured to use the respective supplemental input parameters and the supplemental output parameters corresponding to each supplemental data parameter as respective supplemental task data; sending the supplementary task data to a designated device, displaying the supplementary task data to a designated user through the designated device, and enabling the designated user to determine a confidence level corresponding to the supplementary task data for each supplementary task data, wherein the confidence level is used for characterizing the accuracy of the supplementary output parameters in the supplementary task data obtained by the target task model according to the supplementary input parameters in the supplementary task data; and constructing decision suggestion distribution according to the confidence coefficient and the historical task data.
Optionally, the executing module 405 is specifically configured to construct a confidence coefficient distribution according to the confidence coefficient corresponding to each supplementary task data; constructing decision weight distribution of each supplementary task data according to the confidence coefficient distribution and the expert experience data, wherein the decision weight distribution is used for representing the importance degree of each supplementary task data for optimizing the target task model; and constructing decision suggestion distribution according to the decision weight distribution and the historical task data.
Optionally, the executing module 405 is specifically configured to sample from the value range of the input parameter based on the decision suggestion distribution, to obtain each sample input parameter, and a probability that each sample input parameter is an optimal input parameter; fitting the probability that the input parameter of each sample is the optimal input parameter to obtain an input probability density function; and determining optimal input parameters according to the input probability density function, and executing tasks according to the optimal input parameters.
The present specification also provides a computer readable storage medium storing a computer program operable to perform a task execution method as provided in fig. 1 above.
The present specification also provides a schematic structural diagram of an electronic device corresponding to fig. 1 shown in fig. 5. At the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile storage, as illustrated in fig. 5, although other hardware required by other services may be included. The processor reads the corresponding computer program from the nonvolatile memory into the memory and then runs the computer program to implement the task execution method described in fig. 1. Of course, other implementations, such as logic devices or combinations of hardware and software, are not excluded from the present description, that is, the execution subject of the following processing flows is not limited to each logic unit, but may be hardware or logic devices.
Improvements to one technology can clearly distinguish between improvements in hardware (e.g., improvements to circuit structures such as diodes, transistors, switches, etc.) and software (improvements to the process flow). However, with the development of technology, many improvements of the current method flows can be regarded as direct improvements of hardware circuit structures. Designers almost always obtain corresponding hardware circuit structures by programming improved method flows into hardware circuits. Therefore, an improvement of a method flow cannot be said to be realized by a hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (e.g., field programmable gate array (Field Programmable Gate Array, FPGA)) is an integrated circuit whose logic function is determined by the programming of the device by a user. A designer programs to "integrate" a digital system onto a PLD without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, such programming is mostly implemented by using "logic compiler" software, which is similar to the software compiler used in program development and writing, and the original code before the compiling is also written in a specific programming language, which is called hardware description language (Hardware Description Language, HDL), but not just one of the hdds, but a plurality of kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), lava, lola, myHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog are currently most commonly used. It will also be apparent to those skilled in the art that a hardware circuit implementing the logic method flow can be readily obtained by merely slightly programming the method flow into an integrated circuit using several of the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable logic controllers, and embedded microcontrollers, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic of the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller in a pure computer readable program code, it is well possible to implement the same functionality by logically programming the method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Such a controller may thus be regarded as a kind of hardware component, and means for performing various functions included therein may also be regarded as structures within the hardware component. Or even means for achieving the various functions may be regarded as either software modules implementing the methods or structures within hardware components.
The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being functionally divided into various units, respectively. Of course, the functions of each element may be implemented in one or more software and/or hardware elements when implemented in the present specification.
It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The present description is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
The foregoing is merely exemplary of the present disclosure and is not intended to limit the disclosure. Various modifications and alterations to this specification will become apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, or the like, which are within the spirit and principles of the present description, are intended to be included within the scope of the claims of the present description.
Claims (10)
1. A method of performing a task, comprising:
acquiring a target task model and expert experience data and historical task data corresponding to the target task model, wherein the expert experience data are used for representing the confidence level of data of each dimension of input parameters of the target task model on an output result of the target task model, the target task model is used for executing the target task, the target task model is optimized through an initial task model, the initial task model is constructed through acquired basic data of the target task to be executed, the target task comprises a task for controlling a mechanical arm to conduct operation on a diseased part of a patient, the basic data comprise interaction relation data between various inputtable control instructions of the mechanical arm and excision situation data of the diseased part of the patient corresponding to the control instructions input to the mechanical arm in a historical manner, the inputtable control instructions are used for controlling each joint of the mechanical arm to be adjusted according to the appointed adjustment parameters, and the target task model can recommend a group of instructions for controlling the mechanical arm to conduct operation according to the excision situation data of the diseased part of the patient;
Determining the distribution of the value of each dimension of the historical task data in the value range of the input parameter of the target task model as a value distribution;
sampling from the value range of the input parameters according to the expert experience data and the value distribution to obtain each supplementary input parameter;
inputting the supplementary input parameters into the target task model to obtain supplementary output parameters corresponding to the supplementary input parameters;
and constructing decision suggestion distribution according to the supplementary input parameters, the supplementary output parameters and the historical task data, determining optimal input parameters according to the decision suggestion distribution, and executing tasks according to the optimal input parameters.
2. The method of claim 1, wherein obtaining the target task model and the historical task data comprises:
acquiring an initial task model and historical task data;
acquiring error distribution of the output value of the initial task model as prior error distribution;
according to the error between the output value obtained by inputting the historical task data into the initial task model and the actual output value, optimizing the prior error distribution to obtain posterior error distribution;
And optimizing the initial task model according to the posterior error distribution to obtain a target task model.
3. The method according to claim 1, wherein each supplementary input parameter is sampled from a range of values of said input parameter based on said expert experience data and said value distribution, in particular comprising:
determining the distribution of the weight of each dimension of the input parameters of the target task model according to the expert experience data, wherein the weight of each dimension is used for representing the importance degree of the data of the dimension to the output result of the target task model as an importance distribution;
and sampling from the value range of the input parameters according to the importance distribution and the value distribution to obtain each supplementary input parameter.
4. The method according to claim 1, wherein each supplementary input parameter is sampled from a range of values of said input parameter based on said expert experience data and said value distribution, in particular comprising:
determining a sampling probability density function corresponding to the value range of the input parameter according to the expert experience data and the value distribution, wherein the sampling probability density function is used for representing the probability that each value in the value range of each dimension of the input parameter is sampled;
And sampling from the value range of the input parameters according to the sampling probability density function to obtain each supplementary input parameter.
5. The method of claim 1, wherein constructing a decision suggestion distribution based on the supplemental input parameters, the supplemental output parameters, and the historical task data, comprises:
taking the supplementary input parameters and the supplementary output parameters corresponding to the supplementary data parameters as supplementary task data;
sending the supplementary task data to a designated device, displaying the supplementary task data to a designated user through the designated device, and enabling the designated user to determine a confidence level corresponding to the supplementary task data for each supplementary task data, wherein the confidence level is used for characterizing the accuracy of the supplementary output parameters in the supplementary task data obtained by the target task model according to the supplementary input parameters in the supplementary task data;
and constructing decision suggestion distribution according to the confidence coefficient and the historical task data.
6. The method of claim 5, wherein constructing a decision suggestion distribution based on the confidence level and the historical task data, comprises:
Constructing confidence distribution according to the confidence coefficient corresponding to each supplementary task data;
constructing decision weight distribution of each supplementary task data according to the confidence coefficient distribution and the expert experience data, wherein the decision weight distribution is used for representing the importance degree of each supplementary task data for optimizing the target task model;
and constructing decision suggestion distribution according to the decision weight distribution and the historical task data.
7. The method of claim 6, wherein determining optimal input parameters according to the decision suggestion distribution and performing task execution according to the optimal input parameters, comprises:
sampling from the value range of the input parameters based on the decision suggestion distribution to obtain each sample input parameter and the probability that each sample input parameter is the optimal input parameter;
fitting the probability that the input parameter of each sample is the optimal input parameter to obtain an input probability density function;
and determining optimal input parameters according to the input probability density function, and executing tasks according to the optimal input parameters.
8. A task execution device, characterized by comprising:
The system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring a target task model, expert experience data and historical task data corresponding to the target task model, the expert experience data are used for representing the confidence level of data of each dimension of input parameters of the target task model on an output result of the target task model, the target task model is used for executing the target task, the target task model is obtained through optimization of an initial task model, the initial task model is constructed through acquired basic data of the target task to be executed, the target task comprises a task for controlling a mechanical arm to conduct operation on a diseased part of a patient, the basic data comprise interaction relation data between control instructions which can be input into the mechanical arm and excision situation data of the diseased part of the patient corresponding to the control instructions which are input into the mechanical arm in a historical manner, the control instructions which can be input into the mechanical arm are used for controlling each joint of the mechanical arm to be adjusted according to specified adjustment parameters, and the target task model can recommend a group of instructions for controlling the mechanical arm to conduct operation according to excision situation data of the diseased part of the patient;
A determining module, configured to determine a distribution of values of each dimension of the historical task data in a value range of an input parameter of the target task model as a value distribution;
the first supplementary module is used for sampling each supplementary input parameter from the value range of the input parameter according to the expert experience data and the value distribution;
the second supplementing module is used for inputting the supplementing input parameters into the target task model to obtain the supplementing output parameters corresponding to the supplementing input parameters;
and the execution module is used for constructing decision suggestion distribution according to the supplementary input parameters, the supplementary output parameters and the historical task data, determining optimal input parameters according to the decision suggestion distribution, and executing tasks according to the optimal input parameters.
9. A computer readable storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method of any of the preceding claims 1-7.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of the preceding claims 1-7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311307471.2A CN117057162B (en) | 2023-10-10 | 2023-10-10 | Task execution method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311307471.2A CN117057162B (en) | 2023-10-10 | 2023-10-10 | Task execution method and device, storage medium and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117057162A CN117057162A (en) | 2023-11-14 |
| CN117057162B true CN117057162B (en) | 2024-01-09 |
Family
ID=88662969
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311307471.2A Active CN117057162B (en) | 2023-10-10 | 2023-10-10 | Task execution method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117057162B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109343064A (en) * | 2018-11-23 | 2019-02-15 | 江苏徐工工程机械研究院有限公司 | A kind of mine truck detection system for obstacle and detection method |
| WO2020155788A1 (en) * | 2019-01-30 | 2020-08-06 | 阿里巴巴集团控股有限公司 | Data determination method, apparatus and device, and medium |
| CN116523499A (en) * | 2023-04-28 | 2023-08-01 | 北京航空航天大学 | Automatic fault diagnosis and prediction method and system based on data driving model |
-
2023
- 2023-10-10 CN CN202311307471.2A patent/CN117057162B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109343064A (en) * | 2018-11-23 | 2019-02-15 | 江苏徐工工程机械研究院有限公司 | A kind of mine truck detection system for obstacle and detection method |
| WO2020155788A1 (en) * | 2019-01-30 | 2020-08-06 | 阿里巴巴集团控股有限公司 | Data determination method, apparatus and device, and medium |
| CN116523499A (en) * | 2023-04-28 | 2023-08-01 | 北京航空航天大学 | Automatic fault diagnosis and prediction method and system based on data driving model |
Non-Patent Citations (1)
| Title |
|---|
| 参数互补性目标函数构造方法研究;赵佳宝, 潘韬, 盛昭瀚;管理科学学报(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117057162A (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200211106A1 (en) | Method, apparatus, and device for training risk management models | |
| JP7350878B2 (en) | Image analysis method, device, program | |
| JP6869347B2 (en) | Risk control event automatic processing method and equipment | |
| CN109242519B (en) | Abnormal behavior identification method, device and equipment | |
| US11030411B2 (en) | Methods, apparatuses, and devices for generating word vectors | |
| CN109426912B (en) | Wind control system optimization method, system and device and electronic equipment | |
| CN111401940B (en) | Feature prediction method, device, electronic equipment and storage medium | |
| CN110826894A (en) | Hyper-parameter determination method and device and electronic equipment | |
| US20130080129A1 (en) | Multi-component model engineering | |
| CN116502679B (en) | Model construction method and device, storage medium and electronic equipment | |
| CN116304720B (en) | Cost model training method and device, storage medium and electronic equipment | |
| CN114631099A (en) | Artificial intelligence transparency | |
| CN116230108A (en) | Intelligent medication decision method and system based on TDM and AI technologies | |
| CN107038127A (en) | Application system and its buffer control method and device | |
| CN117057162B (en) | Task execution method and device, storage medium and electronic equipment | |
| CN116524998A (en) | Model training method and molecular property information prediction method and device | |
| CN112927815B (en) | Method, device and equipment for predicting intracranial aneurysm information | |
| US20170123774A1 (en) | Method for defining alias sets | |
| CN108762959B (en) | Method, device and equipment for selecting system parameters | |
| CN113205377A (en) | Information recommendation method and device | |
| WO2024065535A1 (en) | Methods, apparatus, and articles of manufacture to generate hardware-aware machine learning model architectures for multiple domains without training | |
| CN117171577B (en) | Dynamic decision method and device for high-performance operator selection | |
| CN117407690B (en) | Task execution method, device and equipment based on model migration evaluation | |
| CN115827880B (en) | Business execution method and device based on emotion classification | |
| CN117952182B (en) | Mixed precision model training method and device based on data quality |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |