Efficiency index real-time evaluation system, method and program product

Abstract

The invention discloses an efficiency index real-time evaluation system and method and a program product, and belongs to the technical field of artificial intelligence. In the system, a task distributor is configured to distribute efficiency index real-time evaluation tasks to N index real-time evaluation agents, and a decision device is configured to judge the evaluation results of the index real-time evaluation agents and send the correct evaluation results to a comprehensive evaluator for real-time evaluation of the efficiency indexes; evaluating the efficiency index in real time according to the correct evaluation result of the comprehensive evaluator; the advantage value calculator calculates an advantage value according to the evaluation results reported by the N index real-time evaluation agents, the index real-time evaluation functions and the rewards; and the nth index real-time evaluation agent updates own parameters according to the advantage value, and performs efficiency index real-time evaluation through the test data set according to the distributed tasks. The method is high in speed and high in real-time performance.

Description

Technical Field

[0001] The present invention relates to a real-time performance index evaluation system, method and program product, belonging to the technical field of artificial intelligence. Background Art

[0002] The Chinese invention patent application with publication number CN118114873A discloses an intelligent evaluation system for performance indicators, which includes: an indicator system management module for constructing an indicator evaluation system for weapons and equipment to be evaluated; an evaluation scheme management module for screening corresponding marked hierarchical characteristics and the affiliation and evaluation algorithms between each marked hierarchical characteristic, and associating them to obtain a standby evaluation scheme; a data set management module for screening characteristic values ​​from a hierarchical characteristic value set based on the standby evaluation scheme to obtain a standby data set, and associating them to obtain a standby evaluation task; an evaluation task management module for constructing an evaluation standard based on one or more weapons and equipment to be evaluated, and associating the evaluation standard with multiple standby evaluation tasks; an evaluation result generation module for calculating the final evaluation results corresponding to one or more weapons and equipment to be evaluated, and performing visual display.

[0003] Although this invention improves the evaluation efficiency and accuracy, its real-time performance is poor. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a real-time performance indicator evaluation system, method and computer program product, which have high real-time performance.

[0005] To achieve the above-mentioned purpose, the present invention provides a real-time evaluation system for performance indicators, which includes a central processing unit and N real-time evaluation agents, wherein the central processing unit includes a task allocator, a judger, a comprehensive evaluator and an advantage value calculator, wherein the task allocator is configured to allocate real-time evaluation tasks to the N real-time evaluation agents, the judger is configured to judge the evaluation results of the real-time evaluation agents, transmit the correct evaluation results to the comprehensive evaluator for real-time performance evaluation, and discard the wrong evaluation results; the comprehensive evaluator is configured to perform real-time performance evaluation based on the correct evaluation results provided by the N real-time evaluation agents; the advantage value calculator calculates the advantage value based on the evaluation results, evaluation functions and rewards reported by the N real-time evaluation agents; the nth real-time evaluation agent updates its own parameters based on the advantage value, and performs real-time evaluation of the indicators through a test data set based on the assigned tasks, generates evaluation results for the assigned tasks, and reports the evaluation results, evaluation functions and rewards to the central processing unit, n=1,2,…,N.

[0006] To achieve the above-mentioned object of the invention, the present invention provides a real-time performance index evaluation method, which comprises the following steps: Step 1: The task allocator of the central processor allocates the real-time indicator evaluation tasks to N real-time indicator evaluation agents; Step 2: The decision device of the central processing unit makes a decision on the evaluation result of the real-time evaluation agent, transmits the correct evaluation result to the comprehensive evaluator for real-time evaluation of performance indicators, and discards the wrong evaluation result; Step 3: The comprehensive evaluator of the central processor evaluates the correct evaluation results provided by the intelligent agent in real time according to the N indicators to perform real-time evaluation of the indicators; Step 4: The advantage value calculator calculates the advantage value based on the N indicators by evaluating the evaluation results, evaluation functions and rewards reported by the agent in real time; Step 5: The nth indicator real-time evaluation agent updates its own parameters according to the advantage value, and performs real-time evaluation of the performance indicators through the test data set according to the assigned task, generates the evaluation results for the assigned task, and reports the evaluation results, evaluation function and reward to the central processor, n=1,2,…,N.

[0007] To achieve the above object, the present invention further provides a computer program product, which includes computer program code, and the computer program code can be called by a processor to execute the above method.

[0008] Compared with the prior art, the real-time performance index evaluation method, system and computer program product provided by the present invention have the following beneficial effects: The present invention uses trained indicators to evaluate the performance indicators of intelligent agents in real time, with high speed and real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a block diagram of the composition of the performance index real-time evaluation system provided by the first embodiment of the present invention.

[0010] Figure 2 It is a block diagram of the composition of the performance indicator real-time evaluation system provided by the third embodiment of the present invention. DETAILED DESCRIPTION

[0011] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0012] First embodiment

[0013] Figure 1 is a block diagram of the composition of the performance index real-time evaluation system provided by the first embodiment of the present invention, such as Figure 1 As shown, the performance index real-time evaluation system provided by the present invention includes a central processing unit and N real-time index evaluation agents, wherein the central processing unit includes a task allocator, a judger, a comprehensive evaluator and an advantage value calculator, wherein the task allocator is configured to allocate indicator real-time evaluation tasks to the N real-time index evaluation agents, the judger is configured to judge the evaluation results of the real-time index evaluation agents, transmit the correct evaluation results to the comprehensive evaluator for real-time performance evaluation, and discard the wrong evaluation results; the comprehensive evaluator is configured to perform real-time performance evaluation based on the correct evaluation results provided by the N real-time index evaluation agents; the advantage value calculator calculates the advantage value based on the evaluation results, evaluation functions and rewards reported by the N real-time index evaluation agents; the nth real-time index evaluation agent updates its own parameters based on the advantage value, and performs real-time index evaluation through the test data set according to the assigned task, generates an evaluation result for the assigned task, and reports the evaluation result, evaluation function and reward to the central processing unit, n=1,2,…,N.

[0014] The performance indicator real-time evaluation system provided by the first embodiment of the present invention calculates multiple indicator values ​​in parallel through N real-time indicator evaluation agents, thereby speeding up the calculation speed; by setting a judgement device, the performance real-time evaluation is only performed on the correct evaluation results provided by the N real-time indicator evaluation agents, thereby improving the accuracy of the evaluation.

[0015] In the first embodiment, the nth real-time indicator evaluation agent includes a real-time indicator evaluation function and the evaluation function , is the parameter of the real-time evaluation function of the nth real-time evaluation agent, The training of the agent for real-time evaluation of the nth indicator is as follows: The nth indicator evaluates the agent in real time based on the test data Get evaluation results and rewards , and , evaluation function and reward The advantage value calculator reported to the CPU, where ; Advantage Value Calculator calculates the advantage value according to formula 1 , and the advantage value Send to the nth indicator to evaluate the agent in real time: Formula 1: In the formula, is the discount factor; The nth indicator evaluates the agent in real time through Update the parameters of the indicator real-time evaluation function .

[0016] In the first embodiment, the nth indicator real-time evaluation agent is evaluated by the advantage value according to formula 2 Update the parameters of the indicator real-time evaluation function for: Formula 2: , In the formula, Yes The gradient function of is the learning coefficient.

[0017] In the first embodiment, the nth indicator evaluates the agent in real time through the advantage value Update the parameters of the evaluation function .

[0018] In the first embodiment, the nth indicator real-time evaluation agent is evaluated by the advantage value according to Formula 3 Update the parameters of the evaluation function for: Formula 3 .

[0019] The first embodiment of the present invention obtains the advantage value of the above structure, and evaluates the intelligent agent in real time by updating the indicator through the advantage value, thereby accelerating the convergence speed and the training speed.

[0020] In the first embodiment, N real-time indicator evaluation agents respectively provide N indicator values ​​for evaluating the performance of the equipment, and the N indicator values ​​are all transmitted to the comprehensive evaluator to evaluate the performance of the equipment.

[0021] The comprehensive evaluator evaluates the agent in real time based on N indicators to provide N indicator values ​​for real-time performance evaluation, which specifically includes the following steps: S1-1: Generate M evaluation schemes based on the correct evaluation results provided by the intelligent agent in real time according to N indicators to obtain the evaluation scheme matrix : S1-2: Evaluation Scheme Matrix The negative indicators are processed positively to obtain a very large data matrix ; S1-3: Select the maximum number in each column of the extremely large data matrix to form the optimal solution vector ; S1-4: Select the minimum number in each column of the extremely large data matrix to form the worst solution vector ; S1-5: For extremely large data matrices Each element in is normalized to obtain a normalized extremely large data matrix , where ; S1-6: For the mth solution, calculate its distance from 1: ; Calculate its Optimal solution distance: ; S1-7: Calculate the score of the mth solution: ; S1-8: Based on the score Sort and score The highest solution is used as the performance evaluation solution.

[0022] Optionally, step S1-7 and step S1-8 may be replaced by step S1-9 and step S1-10 respectively: S1-9: Calculate the score of the mth solution: ; S1-10: Based on the score Sort and score The smallest solution is used as the performance evaluation solution.

[0023] The comprehensive evaluator provided in the first embodiment of the present invention can obtain the best technical solution for performance evaluation through the above technical solution.

[0024] The performance index real-time evaluation system provided in the first embodiment further includes an input unit, through which a user inputs instructions, test data, etc.

[0025] The performance index real-time evaluation system provided in the first embodiment further includes an output unit, and the output unit includes a display, etc. The display is used to display test data, real-time index evaluation functions, evaluation functions, evaluation results, etc.

[0026] The first embodiment of the present invention performs real-time evaluation of performance indicators through an indicator real-time evaluation agent, which is capable of autonomous learning and has the ability to adapt to new test data and new indicator real-time evaluation tasks.

[0027] The first embodiment of the present invention does not need to use an experience pool to store historical samples, which saves storage space and improves data sampling efficiency, thereby increasing training speed.

[0028] Second embodiment

[0029] The second embodiment of the present invention only describes the contents that are different from the first embodiment, and the same contents will not be described repeatedly.

[0030] A second embodiment of the present invention provides a method for real-time performance index evaluation, which includes the following steps: Step 1: The task allocator of the central processor allocates the real-time indicator evaluation tasks to N real-time indicator evaluation agents; Step 2: The decision device of the central processing unit makes a decision on the evaluation result of the real-time evaluation agent, transmits the correct evaluation result to the comprehensive evaluator for real-time evaluation of performance indicators, and discards the wrong evaluation result; Step 3: The comprehensive evaluator of the central processor evaluates the correct evaluation results provided by the intelligent agent in real time according to the N indicators to perform real-time evaluation of the indicators; Step 4: The advantage value calculator calculates the advantage value based on the N indicators by evaluating the evaluation results, evaluation functions and rewards reported by the agent in real time; Step 5: The nth indicator real-time evaluation agent updates its own parameters according to the advantage value, and performs real-time evaluation of the performance indicators through the test data set according to the assigned task, generates the evaluation results for the assigned task, and reports the evaluation results, evaluation function and reward to the central processor, n=1,2,…,N.

[0031] In the second embodiment of the present invention, the performance index is evaluated in real time by an index real-time evaluation agent, which is capable of autonomous learning and has the ability to adapt to new data resources and new performance index real-time evaluation tasks;

[0032] The second embodiment of the present invention does not need to use an experience pool to store historical samples, which saves storage space and improves data sampling efficiency, thereby increasing training speed. At the same time, by using multiple different training environments to collect samples, the distribution of samples is more uniform, which is more conducive to the training of neural networks.

[0033] Third embodiment

[0034] Figure 2 is a block diagram of a real-time performance indicator evaluation system provided by a third embodiment of the present invention. Figure 2 As shown, the performance indicator real-time evaluation system provided by the present invention includes a central processing unit and N real-time indicator evaluation agents, wherein the central processing unit includes a task allocator, a judger and a comprehensive evaluator, wherein the task allocator is configured to allocate indicator real-time evaluation tasks to the N real-time indicator evaluation agents, the judger is configured to judge the evaluation results of the real-time indicator evaluation agents, transmit the correct evaluation results to the comprehensive evaluator for real-time performance evaluation, and discard the incorrect evaluation results; the comprehensive evaluator is configured to perform real-time performance evaluation based on the correct evaluation results provided by the N real-time indicator evaluation agents.

[0035] The nth indicator real-time evaluation intelligence includes an advantage value calculator, which calculates the advantage value according to the evaluation results, evaluation functions and rewards reported by the nth indicator real-time evaluation agent; the nth indicator real-time evaluation agent updates its own parameters according to the advantage value, and performs real-time evaluation of indicators through a test data set according to the assigned task to generate an evaluation result for the assigned task, n=1,2,…,N.

[0036] The real-time performance evaluation system provided by the third embodiment of the present invention calculates multiple indicator values ​​in parallel through N real-time performance evaluation agents, thereby speeding up the calculation speed; and by setting a judge to only perform real-time performance evaluation on the correct evaluation results provided by the N real-time performance evaluation agents, the accuracy of the evaluation is improved.

[0037] In the third embodiment, the nth real-time indicator evaluation agent includes a real-time indicator evaluation function and the evaluation function , is the parameter of the real-time evaluation function of the nth real-time evaluation agent, , is the parameter of the value function. The training of the real-time evaluation agent for the nth indicator includes the following process: The nth indicator evaluates the agent in real time based on the test data Get evaluation results and rewards ; The advantage value calculator uses formula 1 through , evaluation function and reward Calculate the advantage value ; The nth indicator evaluates the agent in real time through Update the parameters of the indicator real-time evaluation function .

[0038] In the third embodiment, the nth indicator real-time evaluation agent uses the advantage value according to formula 2 Update the parameters of the indicator real-time evaluation function .

[0039] In the third embodiment, the nth indicator evaluates the agent in real time through the advantage value Update the parameters of the evaluation function .

[0040] In the third embodiment, the nth indicator real-time evaluation agent uses the advantage value according to formula 3 Update the parameters of the evaluation function .

[0041] The first embodiment of the present invention obtains the advantage value of the above structure, and evaluates the intelligent agent in real time by updating the indicator through the advantage value, thereby accelerating the convergence speed and the training speed.

[0042] In the first embodiment, N real-time indicator evaluation agents respectively provide N indicator values ​​for evaluating the performance of the equipment, and the N indicator values ​​are all transmitted to the comprehensive evaluator to evaluate the performance of the equipment.

[0043] The comprehensive evaluator provided in the third embodiment of the present invention can obtain the best technical solution for performance evaluation through the above technical solution.

[0044] The performance index real-time evaluation system provided in the third embodiment further includes an input unit, through which the user inputs instructions, test data, etc.

[0045] The performance index real-time evaluation system provided in the third embodiment also includes an output unit, and the output unit includes a display, etc. The display is used to display test data, real-time index evaluation functions, evaluation functions, evaluation results, etc.

[0046] The third embodiment of the present invention performs real-time evaluation of performance indicators through an indicator real-time evaluation agent, which is capable of autonomous learning and has the ability to adapt to new test data and new indicator real-time evaluation tasks.

[0047] The third embodiment of the present invention does not need to use an experience pool to store historical samples, which saves storage space and improves data sampling efficiency, thereby increasing training speed.

[0048] Fourth embodiment

[0049] The fourth embodiment of the present invention only describes the contents that are different from the third embodiment, and the same contents will not be described repeatedly.

[0050] A fourth embodiment of the present invention provides a method for real-time performance index evaluation, which comprises the following steps: S2-1: The task allocator of the central processor allocates the real-time indicator evaluation tasks to N real-time indicator evaluation agents; S2-2: The decision device of the central processing unit makes a decision on the evaluation result of the real-time evaluation agent, transmits the correct evaluation result to the comprehensive evaluation device for real-time evaluation of performance indicators, and discards the wrong evaluation result; S2-3: The comprehensive evaluator of the central processing unit evaluates the correct evaluation results provided by the intelligent agent in real time according to N indicators to perform real-time evaluation of the indicators; S2-4: The nth indicator real-time evaluation agent performs real-time evaluation of the performance indicators according to the assigned task through the test data set, generates an evaluation result for the assigned task, and calculates the advantage value based on the evaluation result, evaluation function and reward, and updates its own parameters n=1,2,…,N according to the advantage value.

[0051] A fourth embodiment of the present invention performs real-time performance index evaluation through an index real-time evaluation agent, which is capable of autonomous learning and has the ability to adapt to new data resources and new performance index real-time evaluation tasks;

[0052] The fourth embodiment of the present invention does not need to use an experience pool to store historical samples, which saves storage space and improves data sampling efficiency, thereby increasing training speed. At the same time, by using multiple different training environments to collect samples, the distribution of samples is more uniform, which is more conducive to the training of neural networks.

[0053] Fifth embodiment

[0054] A fifth embodiment of the present invention provides a computer program product, characterized in that it includes computer program codes, and the computer program codes can be called by a processor to execute the method described in the second embodiment or the fourth embodiment.

[0055] The beneficial effects of the fifth embodiment of the present invention are the same as those of the second embodiment or the fourth embodiment, and will not be described again here.

[0056] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. The meaning of "several" is one or more, unless otherwise clearly and specifically defined.

[0057] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A real-time performance index evaluation system, characterized in that: It includes a central processing unit and N real-time indicator evaluation agents, wherein the central processing unit includes a task distributor, a judger, a comprehensive evaluator and an advantage value calculator, wherein the task distributor is configured to distribute indicator real-time evaluation tasks to the N real-time indicator evaluation agents, the judger is configured to judge the evaluation results of the real-time indicator evaluation agents, transmit the correct evaluation results to the comprehensive evaluator for real-time performance evaluation, and discard the wrong evaluation results; the comprehensive evaluator is configured to perform real-time performance evaluation based on the correct evaluation results provided by the N real-time indicator evaluation agents; the advantage value calculator calculates the advantage value based on the evaluation results, evaluation functions and rewards reported by the N real-time indicator evaluation agents; the nth real-time indicator evaluation agent updates its own parameters based on the advantage value, and performs real-time indicator evaluation through a test data set based on the assigned tasks, generates evaluation results for the assigned tasks, and reports the evaluation results, evaluation functions and rewards to the central processing unit, n=1,2,…,N.

2. The real-time performance evaluation system according to claim 2, characterized in that: The nth indicator real-time evaluation agent includes the indicator real-time evaluation function and the evaluation function , is the parameter of the real-time evaluation function of the nth real-time evaluation agent, The training of the agent for real-time evaluation of the nth indicator is composed of the following steps: The nth indicator evaluates the agent in real time based on the test data Get evaluation results and rewards , and the evaluation results , evaluation function and reward The advantage calculator reports to the CPU, ; The advantage value calculator calculates the advantage value according to the following formula , and the advantage value Send to the nth indicator to evaluate the agent in real time: ， In the formula, is the discount factor; The nth indicator evaluates the agent in real time through Update the parameters of the indicator real-time evaluation function .

3. The real-time performance index evaluation system according to claim 2, characterized in that: The nth indicator evaluates the agent in real time through the advantage value Update the parameters of the indicator real-time evaluation function for: ， In the formula, Yes The gradient function of is the learning coefficient.

4. The real-time performance index evaluation system according to claim 3, characterized in that: The nth indicator evaluates the agent in real time through the advantage value Update the parameters of the evaluation function .

5. The real-time performance index evaluation system according to claim 4, characterized in that: The nth indicator evaluates the agent in real time through the advantage value Update the parameters of the evaluation function for: 。 6. A real-time performance index evaluation method, characterized in that: The steps include: Step 1: The task allocator of the central processor allocates the real-time indicator evaluation tasks to N real-time indicator evaluation agents; Step 2: The decision device of the central processing unit makes a decision on the evaluation result of the real-time evaluation agent, transmits the correct evaluation result to the comprehensive evaluator for real-time evaluation of performance indicators, and discards the wrong evaluation result; Step 3: The comprehensive evaluator of the central processor evaluates the correct evaluation results provided by the intelligent agent in real time according to the N indicators to perform real-time evaluation of the indicators; Step 4: The advantage value calculator calculates the advantage value based on the N indicators by evaluating the evaluation results, evaluation functions and rewards reported by the agent in real time; Step 5: The nth indicator real-time evaluation agent updates its own parameters according to the advantage value, and performs real-time evaluation of the performance indicators through the test data set according to the assigned task, generates the evaluation results for the assigned task, and reports the evaluation results, evaluation function and reward to the central processor, n=1,2,…,N.

7. The real-time performance index evaluation method according to claim 6, characterized in that: The nth indicator real-time evaluation agent includes the indicator real-time evaluation function and the evaluation function , is the parameter of the real-time evaluation function of the nth real-time evaluation agent, The training of the agent for real-time evaluation of the nth indicator is composed of the following steps: The nth indicator evaluates the agent in real time based on the test data Get evaluation results and rewards , and the evaluation results , evaluation function and reward The advantage calculator reports to the CPU, ; The advantage value calculator calculates the advantage value according to the following formula , and the advantage value Transmit to the nth indicator for real-time evaluation of the agent: ， In the formula, is the discount factor; The nth indicator evaluates the agent in real time through Update the parameters of the indicator real-time evaluation function and the parameters of the evaluation function .

8. The real-time performance index evaluation method according to claim 7, characterized in that: The nth indicator evaluates the agent in real time through Update the parameters of the indicator real-time evaluation function Indicator real-time evaluation function parameters for: ， In the formula, Yes The gradient function of is the learning coefficient.

9. The real-time performance index evaluation method according to claim 8, characterized in that: The nth indicator evaluates the agent in real time through the advantage value Update the parameters of the evaluation function for: 。 10. A computer program product, characterized in that It includes computer program code, which can be called by a processor to execute the real-time performance indicator evaluation method described in any one of claims 6-9.

Images