System

Abstract

To provide a system according to an embodiment that quickly selects the optimal generative AI algorithm for a task to which a time limit is set, and efficiently executes the task.SOLUTION: A system according to an embodiment comprises a parallel processing unit, a selection unit, and an acquisition unit. The parallel processing unit simultaneously executes a plurality of generative AI algorithms for a task to which a time limit is set. The selection unit selects a generative AI algorithm on the basis of an execution speed from the plurality of generative AI algorithms executed by the parallel processing unit. The acquisition unit acquires a result of the task executed by the generative AI algorithm selected by the selection unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including a description and related instruction sentence regarding the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] With conventional technologies, it is difficult to quickly select the optimal generative AI algorithm for a task with a time limit, which may hinder efficient task execution.

[0005] The system according to the embodiment aims to quickly select the optimal generative AI algorithm for a task that has a time limit and to execute the task efficiently. [Means for solving the problem]

[0006] The system according to the embodiment includes a parallel processing unit, a selection unit, and an acquisition unit. The parallel processing unit executes multiple generative AI algorithms simultaneously for a task with a time limit set. The selection unit selects a generative AI algorithm from the multiple generative AI algorithms executed by the parallel processing units based on the execution speed. The acquisition unit acquires the execution result of the task by the generative AI algorithm selected by the selection unit. Effect of the Invention

[0007] The system according to the embodiment can quickly select the optimal generative AI algorithm for a task with a time limit and execute the task efficiently. [Brief description of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10]1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] In the following embodiments, a signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0014] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (e.g., a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., voice and / or text). Display 40A displays visible information such as text and images according to instructions from processor 46. Speaker 40B outputs voice according to instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

[0023] FIG. 2 shows an example of main functions of the data processing device 12 and the smart device 14.

[0024] As shown in Fig. 2, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific process program 56 executed on the RAM 30.

[0025] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Also, the data processing device 12 may be a server device, or may be a terminal device (e.g., a mobile phone, a robot, a home appliance, etc.) owned by a user. Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A system according to an embodiment of the present invention is a system that simultaneously executes multiple generative AI algorithms for a task with a time limit and selects the one that can obtain results the fastest. This system uses a parallel processing unit for simultaneously executing multiple generative AI algorithms for a task with a time limit. Next, a selection unit is used to select a generative AI algorithm that can obtain results the fastest based on the execution speed from among the multiple generative AI algorithms executed by the parallel processing unit. A final result is obtained using an acquisition unit that acquires the execution results of the task by the selected generative AI algorithm. Furthermore, an optimization unit is provided that similarly executes multiple generative AI algorithms for multiple tasks and optimizes the selection and execution order of the generative AI algorithm for each task based on the execution results. This makes it possible to optimize the priority order and resource allocation of tasks. For example, by simultaneously executing generative AI algorithms using multiple processors, parallel processing is efficiently performed. In addition, the selection unit selects the optimal generative AI algorithm based on the execution speed of each generative AI algorithm, as well as the accuracy and resource utilization rate. Furthermore, it is possible to monitor the execution results of the selected generative AI algorithm in real time and perform reselection as necessary. The optimization unit includes a machine learning model that learns past execution result data and optimizes the selection of generative AI algorithms and the execution order. This enables more effective schedule management. Finally, it has an interface for visualizing the results of the execution of the generative AI algorithm, allowing users to intuitively understand the results. This allows the system to select the generative AI algorithm that can obtain the fastest results for tasks with time limits, and to process the tasks efficiently.

[0029] The system according to the embodiment includes a parallel processing unit, a selection unit, and an acquisition unit. The parallel processing unit executes a plurality of generating AI algorithms simultaneously for a task with a time limit set. For example, the parallel processing unit can execute the generating AI algorithm simultaneously using a plurality of processors. The selection unit selects a generating AI algorithm based on the execution speed from among the plurality of generating AI algorithms executed by the parallel processing unit. For example, the selection unit can select an optimal generating AI algorithm based on the execution speed of each generating AI algorithm as well as the accuracy and resource usage rate. The acquisition unit acquires the execution result of the task by the generating AI algorithm selected by the selection unit. For example, the acquisition unit can monitor the execution result of the selected generating AI algorithm in real time and perform reselection as necessary. As a result, the system can select a generating AI algorithm that can obtain a result the fastest for a task with a time limit set, and efficiently process the task.

[0030] The parallel processing unit executes multiple generative AI algorithms simultaneously for a task with a time limit. Specifically, the parallel processing unit can utilize multiple processors or cores to execute each generative AI algorithm independently. This allows each algorithm to operate in parallel using its own resources, improving the overall processing speed. For example, the parallel processing unit supports high-speed execution of generative AI algorithms by utilizing dedicated hardware such as GPUs and TPUs. In addition, the parallel processing unit can monitor the execution status of each generative AI algorithm in real time and optimally allocate resources. This minimizes resource waste and achieves efficient processing. Furthermore, the parallel processing unit can provide basic data for selecting the optimal algorithm according to the characteristics of the task, taking into account the characteristics of different generative AI algorithms. For example, if one generative AI algorithm has high accuracy for a certain task, while another algorithm provides fast processing, the parallel processing unit executes these algorithms simultaneously and compares the results to provide information for selecting the optimal algorithm. This allows the parallel processing unit to respond quickly and efficiently to tasks with strict time limits.

[0031] The selection unit selects a generative AI algorithm based on the execution speed from among the multiple generative AI algorithms executed by the parallel processing unit. Specifically, the selection unit measures the execution speed of each generative AI algorithm in real time and selects the optimal algorithm based on the data. For example, the selection unit measures the processing time of each algorithm in milliseconds and selects the algorithm that generates a result in the shortest time. In addition, the selection unit can take into account not only the execution speed but also the accuracy and resource utilization of the generated result. This makes it possible to select an algorithm that is not only fast but also highly accurate and resource-efficient. The selection unit can use a complex evaluation model to comprehensively judge these evaluation criteria and select the optimal generative AI algorithm. For example, the selection unit can use a machine learning model to predict the optimal algorithm based on past execution data. Furthermore, the selection unit can update the selection of algorithms in real time according to the dynamically changing requirements of the task. This allows the selection unit to always select the optimal generative AI algorithm and maximize the efficiency and performance of the entire system.

[0032] The acquisition unit acquires the execution results of the task by the generative AI algorithm selected by the selection unit. Specifically, the acquisition unit can monitor the execution results of the selected generative AI algorithm in real time and perform reselection as necessary. For example, the acquisition unit can periodically check the execution results of the generative AI algorithm, and if the results do not reach the expected quality or accuracy, it can reselect and execute another algorithm. This allows the acquisition unit to respond flexibly to always provide optimal results. Furthermore, the acquisition unit can collect metadata of the execution results and use it for subsequent analysis and evaluation. For example, the acquisition unit collects data such as the generation time, resource usage, and accuracy evaluation of the execution results, and evaluates the performance of the system based on this data. In addition, the acquisition unit can provide an interface for linking the execution results with other systems and modules. This allows the acquisition unit to efficiently manage the execution results of the generative AI algorithm and improve the performance of the entire system. Furthermore, the acquisition unit has a function for saving and backing up the execution results, and supports the protection and restoration of important data. This allows the acquisition unit to realize reliable data management and improve the stability and reliability of the system.

[0033] The system includes an optimization unit that optimizes the selection and execution order of generative AI algorithms for multiple tasks. The optimization unit optimizes the selection and execution order of generative AI algorithms for each task, taking into account execution results of the generative AI algorithms for multiple tasks. For example, the optimization unit includes a machine learning model that learns from past execution result data and optimizes the selection and execution order of generative AI algorithms. This makes it possible to optimize task priorities and resource allocation by optimizing the selection and execution order of generative AI algorithms even for multiple tasks.

[0034] The parallel processing unit can execute the generative AI algorithm simultaneously using multiple processors. The multiple processors include, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and a TPU (Tensor Processing Unit). This allows efficient parallel processing of the generative AI algorithm. For example, the parallel processing unit executes the generative AI algorithm simultaneously using multiple CPUs to improve processing speed. In addition, the parallel processing unit can process large amounts of data at high speed using a GPU. Furthermore, the parallel processing unit can perform training of a machine learning model at high speed using a TPU. This allows the system to execute the generative AI algorithm simultaneously using multiple processors to efficiently process tasks.

[0035] The selection unit can select a generating AI algorithm taking into consideration the accuracy and resource usage rate in addition to the execution speed of each generating AI algorithm. The accuracy includes, for example, accuracy, reproducibility, F1 score, etc. The resource usage rate includes, for example, CPU usage rate, memory usage, etc. This allows the selection unit to select an optimal generating AI algorithm taking into consideration not only the execution speed but also the accuracy and resource usage rate. For example, the selection unit selects a generating AI algorithm with a fast execution speed, high accuracy, and low resource usage rate. The selection unit can also select a generating AI algorithm with a slow execution speed but very high accuracy. Furthermore, the selection unit can also select a generating AI algorithm with a high resource usage rate but very high execution speed and accuracy. This allows the system to select an optimal generating AI algorithm taking into consideration the execution speed, accuracy, and resource usage rate of each generating AI algorithm.

[0036] The selection unit can monitor the execution result of the selected generative AI algorithm in real time and perform reselection as necessary. The selection unit, for example, monitors the execution result of the selected generative AI algorithm in real time and performs reselection as necessary. In order to monitor in real time, for example, it is necessary to set a monitoring tool and a monitoring frequency. As a result, the selection unit can select a more appropriate generative AI algorithm by monitoring the execution result in real time and performing reselection as necessary. For example, the selection unit reselects another generative AI algorithm when the execution result is not as expected. In addition, the selection unit can also perform reselection when the execution result changes midway. Furthermore, the selection unit can also perform reselection when the execution result is updated in real time. As a result, the system can select a more appropriate generative AI algorithm by monitoring the execution result of the selected generative AI algorithm in real time and performing reselection as necessary.

[0037] The optimization unit includes a machine learning model that learns based on past execution result data and selects a generating AI algorithm and optimizes the execution order. The machine learning model includes, for example, a neural network, a decision tree, and a support vector machine. As a result, the optimization unit selects a generating AI algorithm and optimizes the execution order by learning from the past execution result data. For example, the optimization unit analyzes the past execution result data and selects the most efficient generating AI algorithm. The optimization unit can also select a generating AI algorithm with low resource usage from the past execution result data. Furthermore, the optimization unit can select a generating AI algorithm with a low error rate based on the past execution result data. As a result, the system selects a generating AI algorithm and optimizes the execution order by learning from the past execution result data, enabling efficient task processing.

[0038] The system includes an interface for visualizing an execution result of the generative AI algorithm. The interface is, for example, for visualizing the execution result of the generative AI algorithm. The visualization includes, for example, a graph, a chart, a dashboard, and the like. In this way, the interface visualizes the execution result of the generative AI algorithm, thereby enabling the user to intuitively understand the result. For example, the interface displays the execution result as a graph, allowing the user to grasp the trend of the result at a glance. In addition, the interface displays the execution result as a chart, allowing the user to check the details of the result. In addition, the interface displays the execution result as a dashboard, allowing the user to compare multiple results simultaneously. In this way, the system includes an interface for visualizing the execution result of the generative AI algorithm, thereby enabling the user to intuitively understand the result.

[0039] During parallel processing, the parallel processing unit can optimally allocate processors by referring to past execution results of each generating AI algorithm. Past execution results include, for example, log data and history data. As a result, the parallel processing unit can optimally allocate processors by referring to past execution results, enabling efficient parallel processing. For example, the parallel processing unit analyzes past execution times of each generating AI algorithm and allocates them to the most efficient processor. The parallel processing unit can also allocate processors by referring to past resource usage rates of each generating AI algorithm and taking into account resource balance. Furthermore, the parallel processing unit can also allocate processors to highly reliable processors by taking into account past error rates of each generating AI algorithm. As a result, the system can optimally allocate processors by referring to past execution results, enabling efficient parallel processing.

[0040] The parallel processing unit can apply different parallel processing algorithms depending on the type of task during parallel processing. The types of tasks include, for example, data processing tasks, image recognition tasks, and natural language processing tasks. As a result, the parallel processing unit can apply an optimal parallel processing algorithm depending on the type of task, thereby enabling efficient processing. For example, the parallel processing unit applies a high-performance parallel processing algorithm to a task that requires a large amount of calculation. In addition, the parallel processing unit can also apply a parallel processing algorithm with high communication efficiency to a task that requires a large amount of data transfer. Furthermore, the parallel processing unit can also apply a low-latency parallel processing algorithm to a task that requires real-time performance. As a result, the system can apply an optimal parallel processing algorithm depending on the type of task, thereby enabling efficient processing.

[0041] The parallel processing unit can allocate processors during parallel processing, taking into account the geographical distribution of tasks. The geographical distribution includes, for example, communication delays in each region and the location of data centers. As a result, the parallel processing unit can allocate processors while minimizing communication delays by taking into account the geographical distribution of tasks. For example, the parallel processing unit allocates tasks to processors that are geographically close, thereby minimizing communication delays. The parallel processing unit can also allocate tasks to geographically distributed processors to distribute the load. Furthermore, the parallel processing unit can assign tasks to processors located in geographically important locations, thereby prioritizing processing in a specific region. As a result, the system can allocate processors while minimizing communication delays by taking into account the geographical distribution of tasks.

[0042] During parallel processing, the parallel processing unit can select an optimal parallel processing method by referring to documents related to the task. The relevant documents include, for example, academic papers, technical reports, patent documents, and the like. As a result, the parallel processing unit can select an optimal parallel processing method by referring to documents related to the task, enabling efficient processing. For example, the parallel processing unit can select an optimal parallel processing method by referring to the latest research papers related to the task. The parallel processing unit can also select an optimal parallel processing method by referring to patent documents related to the task. Furthermore, the parallel processing unit can also select an optimal parallel processing method by referring to technical documents related to the task. As a result, the system can select an optimal parallel processing method by referring to documents related to the task, enabling efficient processing.

[0043] At the time of selection, the selection unit can improve the accuracy of selection by referring to past execution results of each generating AI algorithm. Past execution results include, for example, log data and history data. As a result, the selection unit can improve the accuracy of selection by referring to past execution results and select the optimal generating AI algorithm. For example, the selection unit analyzes the past execution time of each generating AI algorithm and selects the most efficient algorithm. The selection unit can also refer to the past accuracy of each generating AI algorithm and select the most reliable algorithm. Furthermore, the selection unit can take into account the past resource usage rate of each generating AI algorithm and select the most balanced algorithm. As a result, the system can improve the accuracy of selection by referring to past execution results and select the optimal generating AI algorithm.

[0044] The selection unit can apply different selection algorithms depending on the type of task at the time of selection. Examples of the selection algorithm include random forest and ensemble learning. This allows the selection unit to make an efficient selection by applying an optimal selection algorithm depending on the type of task. For example, the selection unit applies a high-speed selection algorithm to a task that requires a large amount of calculation. The selection unit can also apply a selection algorithm with high communication efficiency to a task that requires a large amount of data transfer. Furthermore, the selection unit can also apply a low-latency selection algorithm to a task that requires real-time performance. This allows the system to make an efficient selection by applying an optimal selection algorithm depending on the type of task.

[0045] At the time of selection, the selection unit can select a generation AI algorithm taking into consideration the geographic distribution of tasks. The geographic distribution includes, for example, communication delays by region and the location of data centers. As a result, the selection unit can select a generation AI algorithm that minimizes communication delays by taking into consideration the geographic distribution of tasks. For example, the selection unit preferentially selects a generation AI algorithm that is geographically close, thereby minimizing communication delays. The selection unit can also select a generation AI algorithm that is geographically distributed, thereby balancing the load. Furthermore, the selection unit can select a generation AI algorithm that is located in a geographically important location, thereby prioritizing processing in a specific region. As a result, the system can select a generation AI algorithm that minimizes communication delays by taking into consideration the geographic distribution of tasks.

[0046] The selection unit can improve the accuracy of the selection by referring to the related literature of the task at the time of selection. The related literature includes, for example, academic papers, technical reports, patent documents, etc. As a result, the selection unit can improve the accuracy of the selection by referring to the related literature of the task and select the optimal generative AI algorithm. For example, the selection unit refers to the latest research papers related to the task and selects the optimal generative AI algorithm. The selection unit can also refer to patent documents related to the task and select the optimal generative AI algorithm. Furthermore, the selection unit can also refer to technical documents related to the task and select the optimal generative AI algorithm. As a result, the system can improve the accuracy of the selection by referring to the related literature of the task and select the optimal generative AI algorithm.

[0047] The acquisition unit can select the optimal acquisition method by referring to the past execution results of each generating AI algorithm at the time of acquisition. The past execution results include, for example, log data and history data. As a result, the acquisition unit can select the optimal acquisition method by referring to the past execution results, thereby enabling efficient acquisition. For example, the acquisition unit analyzes the past execution time of each generating AI algorithm and selects the most efficient acquisition method. The acquisition unit can also select the most reliable acquisition method by referring to the past accuracy of each generating AI algorithm. Furthermore, the acquisition unit can also select the most balanced acquisition method by taking into account the past resource usage rate of each generating AI algorithm. As a result, the system can select the optimal acquisition method by referring to the past execution results, thereby enabling efficient acquisition.

[0048] The acquisition unit can apply different acquisition algorithms depending on the type of task at the time of acquisition. The acquisition algorithms include, for example, scraping and data mining. As a result, the acquisition unit can perform efficient acquisition by applying the optimal acquisition algorithm depending on the type of task. For example, the acquisition unit applies a high-speed acquisition algorithm to a task with a large amount of calculation. In addition, the acquisition unit can also apply an acquisition algorithm with high communication efficiency to a task with a large amount of data transfer. Furthermore, the acquisition unit can also apply a low-latency acquisition algorithm to a task that requires real-time performance. As a result, the system can perform efficient acquisition by applying the optimal acquisition algorithm depending on the type of task.

[0049] The acquisition unit can acquire the execution results by taking into consideration the geographical distribution of the tasks. The geographical distribution includes, for example, communication delays in each region and the location of the data center. As a result, the acquisition unit can acquire the execution results with the communication delays minimized by taking into consideration the geographical distribution of the tasks. For example, the acquisition unit preferentially acquires the execution results of the generation AI algorithms that are geographically close, thereby minimizing the communication delays. The acquisition unit can also acquire the execution results of the generation AI algorithms that are geographically distributed, thereby distributing the load. Furthermore, the acquisition unit can acquire the execution results of the generation AI algorithms that are located in geographically important locations, thereby prioritizing processing in a specific region. As a result, the system can acquire the execution results with the communication delays minimized by taking into consideration the geographical distribution of the tasks.

[0050] The acquisition unit can improve the accuracy of acquisition by referring to the related literature of the task at the time of acquisition. The related literature includes, for example, academic papers, technical reports, patent documents, etc. As a result, the acquisition unit can improve the accuracy of acquisition and select the optimal acquisition method by referring to the related literature of the task. For example, the acquisition unit refers to the latest research paper related to the task and selects the optimal acquisition method. The acquisition unit can also refer to patent documents related to the task and select the optimal acquisition method. Furthermore, the acquisition unit can also refer to technical documents related to the task and select the optimal acquisition method. As a result, the system can improve the accuracy of acquisition and select the optimal acquisition method by referring to the related literature of the task.

[0051] The optimization unit can optimize the optimization algorithm by referring to past execution result data during optimization. For example, the optimization unit optimizes the optimization algorithm by referring to past execution result data during optimization. The past execution result data includes, for example, log data and history data. As a result, the optimization unit improves the accuracy of the optimization algorithm by referring to the past execution result data, and efficient optimization is possible. For example, the optimization unit analyzes the past execution result data and selects the most efficient optimization algorithm. The optimization unit can also select an optimization algorithm with low resource usage from the past execution result data. Furthermore, the optimization unit can select an optimization algorithm with a low error rate based on the past execution result data. As a result, the system improves the accuracy of the optimization algorithm by referring to the past execution result data, and efficient optimization is possible.

[0052] The optimization unit can apply different optimization algorithms depending on the type of task during optimization. Examples of optimization algorithms include genetic algorithms and simulated annealing. As a result, the optimization unit can apply an optimal optimization algorithm depending on the type of task, thereby enabling efficient optimization. For example, the optimization unit applies a high-speed optimization algorithm to a task with a large amount of calculations. In addition, the optimization unit can also apply an optimization algorithm with high communication efficiency to a task with a large amount of data transfer. Furthermore, the optimization unit can also apply a low-latency optimization algorithm to a task that requires real-time performance. As a result, the system can apply an optimal optimization algorithm depending on the type of task, thereby enabling efficient optimization.

[0053] The optimization unit can perform optimization by taking into account the geographical distribution of tasks. The geographical distribution includes, for example, communication delays by region and the location of data centers. As a result, the optimization unit can perform optimization with minimized communication delays by taking into account the geographical distribution of tasks. For example, the optimization unit prioritizes optimization of geographically close tasks to minimize communication delays. The optimization unit can also optimize geographically distributed tasks to distribute the load. Furthermore, the optimization unit can optimize tasks located in geographically important locations to prioritize processing in specific regions. As a result, the system can perform optimization with minimized communication delays by taking into account the geographical distribution of tasks.

[0054] The optimization unit can improve the accuracy of optimization by referring to the related literature of the task during optimization. The related literature includes, for example, academic papers, technical reports, patent documents, and the like. As a result, the optimization unit can improve the accuracy of optimization and enable efficient optimization by referring to the related literature of the task. For example, the optimization unit can select an optimal optimization method by referring to the latest research papers related to the task. The optimization unit can also select an optimal optimization method by referring to patent documents related to the task. Furthermore, the optimization unit can also select an optimal optimization method by referring to technical documents related to the task. As a result, the system can improve the accuracy of optimization and enable efficient optimization by referring to the related literature of the task.

[0055] When displaying the interface, the interface can select the optimal display method by referring to the user's past operation history. The operation history includes, for example, a click history and an operation log. This allows the interface to select the optimal display method by referring to the past operation history, enabling efficient display. For example, the interface preferentially provides a display method that the user has frequently used in the past. The interface can also suggest the most efficient display method from the user's past operation history. Furthermore, the interface can analyze the user's past operation history and select a display method with high visibility. This allows the system to select the optimal display method by referring to the past operation history, enabling efficient display.

[0056] The interface can select an optimal display method in consideration of device information of the user when displaying the interface. For example, the interface selects an optimal display method in consideration of device information of the user when displaying the interface. The device information includes, for example, the type of device, the screen size, and the OS version. As a result, the interface selects an optimal display method in consideration of the device information of the user, and efficient display is possible. For example, when the user is using a smartphone, the interface provides a display method that matches the screen size. In addition, when the user is using a tablet, the interface can also provide a display method that is optimized for a large screen. Furthermore, when the user is using a smart watch, the interface can also provide a simple and highly visible display method. As a result, the system selects an optimal display method in consideration of device information of the user, and efficient display is possible.

[0057] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0058] The system can monitor resource usage in real time while a task is being executed and optimally allocate resources. For example, if a particular generative AI algorithm is consuming excessive resources, the system will reallocate resources to other algorithms. Also, if resource usage is low, the system can effectively utilize resources by executing additional tasks. Furthermore, depending on resource usage, the system can dynamically change task priorities to achieve efficient task processing. This allows the system to optimally allocate resources and enable efficient task management.

[0059] The system can detect errors in real time while a task is being performed and take measures to minimize the impact of the error. For example, if a particular generative AI algorithm generates an error, the system can switch to another algorithm. Also, if the frequency of errors is high, the system can identify the cause of the error and provide feedback to correct it. Furthermore, to minimize the impact of errors, the system can automatically rerun the task. This allows the system to minimize the impact of errors and enable efficient task processing.

[0060] The system can check the consistency of data in real time while a task is being executed, and automatically correct any inconsistencies that occur. For example, if a data inconsistency is detected, the system automatically corrects the data and continues executing the task. In addition, if data inconsistencies occur frequently, the system can identify the cause of the inconsistency and provide feedback to correct it. Furthermore, to maintain data consistency, the system can periodically check the data and make corrections as necessary. This allows the system to maintain data consistency and enable efficient task processing.

[0061] The system can monitor performance in real time while a task is being executed and optimize performance. For example, if the performance of a particular generative AI algorithm degrades, the system will switch to another algorithm. Also, if performance degradation occurs frequently, the system can identify the cause of the performance degradation and provide feedback to correct it. Furthermore, to optimize performance, the system can periodically check performance and optimize as necessary. This allows the system to optimize performance and enable efficient task processing.

[0062] The system can detect security risks in real time while a task is being performed and take measures to minimize the impact of the risk. For example, if a specific generative AI algorithm generates a security risk, the system can switch to another algorithm. Also, if the frequency of security risks is high, the system can identify the cause of the risk and provide feedback to correct it. Furthermore, to minimize the impact of security risks, the system can automatically rerun the task. This allows the system to minimize the impact of security risks and enable efficient task processing.

[0063] The process flow of the first embodiment will be briefly described below.

[0064] Step 1: The parallel processing unit executes multiple generative AI algorithms simultaneously for a task with a time limit. For example, the parallel processing unit can execute the generative AI algorithms simultaneously using multiple processors. Step 2: The selection unit selects a generation AI algorithm from among the multiple generation AI algorithms executed by the parallel processing unit based on the execution speed. For example, the selection unit can select an optimal generation AI algorithm based on the execution speed of each generation AI algorithm as well as accuracy and resource usage. Step 3: The acquisition unit acquires the execution result of the task by the generation AI algorithm selected by the selection unit. For example, the acquisition unit can monitor the execution result of the selected generation AI algorithm in real time and perform reselection as necessary.

[0065] (Example 2) A system according to an embodiment of the present invention is a system that simultaneously executes multiple generative AI algorithms for a task with a time limit and selects the one that can obtain results the fastest. This system uses a parallel processing unit for simultaneously executing multiple generative AI algorithms for a task with a time limit. Next, a selection unit is used to select a generative AI algorithm that can obtain results the fastest based on the execution speed from among the multiple generative AI algorithms executed by the parallel processing unit. A final result is obtained using an acquisition unit that acquires the execution results of the task by the selected generative AI algorithm. Furthermore, an optimization unit is provided that similarly executes multiple generative AI algorithms for multiple tasks and optimizes the selection and execution order of the generative AI algorithm for each task based on the execution results. This makes it possible to optimize the priority order and resource allocation of tasks. For example, by simultaneously executing generative AI algorithms using multiple processors, parallel processing is efficiently performed. In addition, the selection unit selects the optimal generative AI algorithm based on the execution speed of each generative AI algorithm, as well as the accuracy and resource utilization rate. Furthermore, it is possible to monitor the execution results of the selected generative AI algorithm in real time and perform reselection as necessary. The optimization unit includes a machine learning model that learns past execution result data and optimizes the selection of generative AI algorithms and the execution order. This enables more effective schedule management. Finally, it has an interface for visualizing the results of the execution of the generative AI algorithm, allowing users to intuitively understand the results. This allows the system to select the generative AI algorithm that can obtain the fastest results for tasks with time limits, and to process the tasks efficiently.

[0066] The system according to the embodiment includes a parallel processing unit, a selection unit, and an acquisition unit. The parallel processing unit executes a plurality of generating AI algorithms simultaneously for a task with a time limit set. For example, the parallel processing unit can execute the generating AI algorithm simultaneously using a plurality of processors. The selection unit selects a generating AI algorithm based on the execution speed from among the plurality of generating AI algorithms executed by the parallel processing unit. For example, the selection unit can select an optimal generating AI algorithm based on the execution speed of each generating AI algorithm as well as the accuracy and resource usage rate. The acquisition unit acquires the execution result of the task by the generating AI algorithm selected by the selection unit. For example, the acquisition unit can monitor the execution result of the selected generating AI algorithm in real time and perform reselection as necessary. As a result, the system can select a generating AI algorithm that can obtain a result the fastest for a task with a time limit set, and efficiently process the task.

[0067] The parallel processing unit executes multiple generative AI algorithms simultaneously for a task with a time limit. Specifically, the parallel processing unit can utilize multiple processors or cores to execute each generative AI algorithm independently. This allows each algorithm to operate in parallel using its own resources, improving the overall processing speed. For example, the parallel processing unit supports high-speed execution of generative AI algorithms by utilizing dedicated hardware such as GPUs and TPUs. In addition, the parallel processing unit can monitor the execution status of each generative AI algorithm in real time and optimally allocate resources. This minimizes resource waste and achieves efficient processing. Furthermore, the parallel processing unit can provide basic data for selecting the optimal algorithm according to the characteristics of the task, taking into account the characteristics of different generative AI algorithms. For example, if one generative AI algorithm has high accuracy for a certain task, while another algorithm provides fast processing, the parallel processing unit executes these algorithms simultaneously and compares the results to provide information for selecting the optimal algorithm. This allows the parallel processing unit to respond quickly and efficiently to tasks with strict time limits.

[0068] The selection unit selects a generative AI algorithm based on the execution speed from among the multiple generative AI algorithms executed by the parallel processing unit. Specifically, the selection unit measures the execution speed of each generative AI algorithm in real time and selects the optimal algorithm based on the data. For example, the selection unit measures the processing time of each algorithm in milliseconds and selects the algorithm that generates a result in the shortest time. In addition, the selection unit can take into account not only the execution speed but also the accuracy and resource utilization of the generated result. This makes it possible to select an algorithm that is not only fast but also highly accurate and resource-efficient. The selection unit can use a complex evaluation model to comprehensively judge these evaluation criteria and select the optimal generative AI algorithm. For example, the selection unit can use a machine learning model to predict the optimal algorithm based on past execution data. Furthermore, the selection unit can update the selection of algorithms in real time according to the dynamically changing requirements of the task. This allows the selection unit to always select the optimal generative AI algorithm and maximize the efficiency and performance of the entire system.

[0069] The acquisition unit acquires the execution results of the task by the generative AI algorithm selected by the selection unit. Specifically, the acquisition unit can monitor the execution results of the selected generative AI algorithm in real time and perform reselection as necessary. For example, the acquisition unit can periodically check the execution results of the generative AI algorithm, and if the results do not reach the expected quality or accuracy, it can reselect and execute another algorithm. This allows the acquisition unit to respond flexibly to always provide optimal results. Furthermore, the acquisition unit can collect metadata of the execution results and use it for subsequent analysis and evaluation. For example, the acquisition unit collects data such as the generation time, resource usage, and accuracy evaluation of the execution results, and evaluates the performance of the system based on this data. In addition, the acquisition unit can provide an interface for linking the execution results with other systems and modules. This allows the acquisition unit to efficiently manage the execution results of the generative AI algorithm and improve the performance of the entire system. Furthermore, the acquisition unit has a function for saving and backing up the execution results, and supports the protection and restoration of important data. This allows the acquisition unit to realize reliable data management and improve the stability and reliability of the system.

[0070] The system includes an optimization unit that optimizes the selection and execution order of generative AI algorithms for multiple tasks. The optimization unit optimizes the selection and execution order of generative AI algorithms for each task, taking into account execution results of the generative AI algorithms for multiple tasks. For example, the optimization unit includes a machine learning model that learns from past execution result data and optimizes the selection and execution order of generative AI algorithms. This makes it possible to optimize task priorities and resource allocation by optimizing the selection and execution order of generative AI algorithms even for multiple tasks.

[0071] The parallel processing unit can execute the generative AI algorithm simultaneously using multiple processors. The multiple processors include, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and a TPU (Tensor Processing Unit). This allows efficient parallel processing of the generative AI algorithm. For example, the parallel processing unit executes the generative AI algorithm simultaneously using multiple CPUs to improve processing speed. In addition, the parallel processing unit can process large amounts of data at high speed using a GPU. Furthermore, the parallel processing unit can perform training of a machine learning model at high speed using a TPU. This allows the system to execute the generative AI algorithm simultaneously using multiple processors to efficiently process tasks.

[0072] The selection unit can select a generating AI algorithm taking into consideration the accuracy and resource usage rate in addition to the execution speed of each generating AI algorithm. The accuracy includes, for example, accuracy, reproducibility, F1 score, etc. The resource usage rate includes, for example, CPU usage rate, memory usage, etc. This allows the selection unit to select an optimal generating AI algorithm taking into consideration not only the execution speed but also the accuracy and resource usage rate. For example, the selection unit selects a generating AI algorithm with a fast execution speed, high accuracy, and low resource usage rate. The selection unit can also select a generating AI algorithm with a slow execution speed but very high accuracy. Furthermore, the selection unit can also select a generating AI algorithm with a high resource usage rate but very high execution speed and accuracy. This allows the system to select an optimal generating AI algorithm taking into consideration the execution speed, accuracy, and resource usage rate of each generating AI algorithm.

[0073] The selection unit can monitor the execution result of the selected generative AI algorithm in real time and perform reselection as necessary. In order to monitor in real time, for example, it is necessary to set a monitoring tool and a monitoring frequency. As a result, the selection unit can select a more appropriate generative AI algorithm by monitoring the execution result in real time and performing reselection as necessary. For example, the selection unit reselects another generative AI algorithm when the execution result is not as expected. In addition, the selection unit can also perform reselection when the execution result changes midway. Furthermore, the selection unit can also perform reselection when the execution result is updated in real time. As a result, the system can select a more appropriate generative AI algorithm by monitoring the execution result of the selected generative AI algorithm in real time and performing reselection as necessary.

[0074] The optimization unit includes a machine learning model that learns based on past execution result data and selects a generating AI algorithm and optimizes the execution order. The machine learning model includes, for example, a neural network, a decision tree, and a support vector machine. As a result, the optimization unit selects a generating AI algorithm and optimizes the execution order by learning from the past execution result data. For example, the optimization unit analyzes the past execution result data and selects the most efficient generating AI algorithm. The optimization unit can also select a generating AI algorithm with low resource usage from the past execution result data. Furthermore, the optimization unit can select a generating AI algorithm with a low error rate based on the past execution result data. As a result, the system selects a generating AI algorithm and optimizes the execution order by learning from the past execution result data, enabling efficient task processing.

[0075] The system includes an interface for visualizing an execution result of the generative AI algorithm. The interface is, for example, for visualizing the execution result of the generative AI algorithm. The visualization includes, for example, a graph, a chart, a dashboard, and the like. In this way, the interface visualizes the execution result of the generative AI algorithm, thereby enabling the user to intuitively understand the result. For example, the interface displays the execution result as a graph, allowing the user to grasp the trend of the result at a glance. In addition, the interface displays the execution result as a chart, allowing the user to check the details of the result. In addition, the interface displays the execution result as a dashboard, allowing the user to compare multiple results simultaneously. In this way, the system includes an interface for visualizing the execution result of the generative AI algorithm, thereby enabling the user to intuitively understand the result.

[0076] The parallel processing unit can estimate the user's emotion and adjust the start timing of the parallel processing in consideration of the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is stressed, the start timing of the parallel processing is delayed to wait until the user is relaxed. Also, when the user is relaxed, the parallel processing is started immediately to efficiently process the task. Furthermore, when the user is in a hurry, the parallel processing is started immediately and adjusted to obtain the result as quickly as possible. As a result, the system can start the parallel processing at a more appropriate timing by adjusting the start timing of the parallel processing according to the user's emotion.

[0077] During parallel processing, the parallel processing unit can optimally allocate processors by referring to past execution results of each generating AI algorithm. Past execution results include, for example, log data and history data. As a result, the parallel processing unit can optimally allocate processors by referring to past execution results, enabling efficient parallel processing. For example, the parallel processing unit analyzes past execution times of each generating AI algorithm and allocates them to the most efficient processor. The parallel processing unit can also allocate processors by referring to past resource usage rates of each generating AI algorithm and taking into account resource balance. Furthermore, the parallel processing unit can also allocate processors to highly reliable processors by taking into account past error rates of each generating AI algorithm. As a result, the system can optimally allocate processors by referring to past execution results, enabling efficient parallel processing.

[0078] The parallel processing unit can apply different parallel processing algorithms depending on the type of task during parallel processing. The types of tasks include, for example, data processing tasks, image recognition tasks, and natural language processing tasks. As a result, the parallel processing unit can apply an optimal parallel processing algorithm depending on the type of task, thereby enabling efficient processing. For example, the parallel processing unit applies a high-performance parallel processing algorithm to a task that requires a large amount of calculation. In addition, the parallel processing unit can also apply a parallel processing algorithm with high communication efficiency to a task that requires a large amount of data transfer. Furthermore, the parallel processing unit can also apply a low-latency parallel processing algorithm to a task that requires real-time performance. As a result, the system can apply an optimal parallel processing algorithm depending on the type of task, thereby enabling efficient processing.

[0079] The parallel processing unit can estimate the user's emotion and determine the priority of parallel processing in consideration of the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is stressed, a task with low importance is processed preferentially to reduce the burden on the user. Also, when the user is relaxed, a task with high importance is processed preferentially to complete the task efficiently. Furthermore, when the user is in a hurry, the most time-consuming task is processed preferentially to shorten the overall processing time. As a result, the system can process tasks with more appropriate priority by determining the priority of parallel processing according to the user's emotion.

[0080] The parallel processing unit can allocate processors in consideration of the geographical distribution of tasks during parallel processing. The parallel processing unit, for example, allocates processors in consideration of the geographical distribution of tasks during parallel processing. The geographical distribution includes, for example, communication delays for each region and the location of data centers. In this way, the parallel processing unit can allocate processors while minimizing communication delays by considering the geographical distribution of tasks. For example, the parallel processing unit allocates tasks to processors that are geographically close, thereby minimizing communication delays. The parallel processing unit can also allocate tasks to geographically distributed processors to distribute loads. Furthermore, the parallel processing unit can also allocate tasks to processors located in geographically important locations, thereby prioritizing processing in a specific region. In this way, the system can allocate processors while minimizing communication delays by considering the geographical distribution of tasks.

[0081] During parallel processing, the parallel processing unit can select an optimal parallel processing method by referring to documents related to the task. The relevant documents include, for example, academic papers, technical reports, patent documents, and the like. As a result, the parallel processing unit can select an optimal parallel processing method by referring to documents related to the task, enabling efficient processing. For example, the parallel processing unit can select an optimal parallel processing method by referring to the latest research papers related to the task. The parallel processing unit can also select an optimal parallel processing method by referring to patent documents related to the task. Furthermore, the parallel processing unit can also select an optimal parallel processing method by referring to technical documents related to the task. As a result, the system can select an optimal parallel processing method by referring to documents related to the task, enabling efficient processing.

[0082] The selection unit can estimate the user's emotion and adjust the selection criteria of the generation AI algorithm taking into account the estimated user's emotion. The estimation of emotion is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is stressed, a generation AI algorithm that emphasizes speed over accuracy is selected. Also, when the user is relaxed, a generation AI algorithm that emphasizes accuracy over speed is selected. Furthermore, when the user is in a hurry, a generation AI algorithm that minimizes resource usage is selected. This allows the system to select a more appropriate algorithm by adjusting the selection criteria of the generation AI algorithm according to the user's emotion.

[0083] At the time of selection, the selection unit can improve the accuracy of selection by referring to past execution results of each generating AI algorithm. Past execution results include, for example, log data and history data. As a result, the selection unit can improve the accuracy of selection by referring to past execution results and select the optimal generating AI algorithm. For example, the selection unit analyzes the past execution time of each generating AI algorithm and selects the most efficient algorithm. The selection unit can also refer to the past accuracy of each generating AI algorithm and select the most reliable algorithm. Furthermore, the selection unit can take into account the past resource usage rate of each generating AI algorithm and select the most balanced algorithm. As a result, the system can improve the accuracy of selection by referring to past execution results and select the optimal generating AI algorithm.

[0084] The selection unit can apply different selection algorithms depending on the type of task at the time of selection. Examples of the selection algorithm include random forest and ensemble learning. This allows the selection unit to make an efficient selection by applying an optimal selection algorithm depending on the type of task. For example, the selection unit applies a high-speed selection algorithm to a task that requires a large amount of calculation. The selection unit can also apply a selection algorithm with high communication efficiency to a task that requires a large amount of data transfer. Furthermore, the selection unit can also apply a low-latency selection algorithm to a task that requires real-time performance. This allows the system to make an efficient selection by applying an optimal selection algorithm depending on the type of task.

[0085] The selection unit can estimate the user's emotion and adjust the selection order of the generation AI algorithms in consideration of the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is stressed, the generation AI algorithms are selected in order from the easiest task. Also, when the user is relaxed, the generation AI algorithms are selected in order from the most complex task. Furthermore, when the user is in a hurry, the generation AI algorithms are selected in order from the most time-consuming task. In this way, the system can select algorithms in a more appropriate order by adjusting the selection order of the generation AI algorithms according to the user's emotion.

[0086] At the time of selection, the selection unit can select a generation AI algorithm taking into consideration the geographic distribution of tasks. The geographic distribution includes, for example, communication delays by region and the location of data centers. As a result, the selection unit can select a generation AI algorithm that minimizes communication delays by taking into consideration the geographic distribution of tasks. For example, the selection unit preferentially selects a generation AI algorithm that is geographically close, thereby minimizing communication delays. The selection unit can also select a generation AI algorithm that is geographically distributed, thereby balancing the load. Furthermore, the selection unit can select a generation AI algorithm that is located in a geographically important location, thereby prioritizing processing in a specific region. As a result, the system can select a generation AI algorithm that minimizes communication delays by taking into consideration the geographic distribution of tasks.

[0087] The selection unit can improve the accuracy of the selection by referring to the related literature of the task at the time of selection. The related literature includes, for example, academic papers, technical reports, patent documents, etc. As a result, the selection unit can improve the accuracy of the selection by referring to the related literature of the task and select the optimal generative AI algorithm. For example, the selection unit refers to the latest research papers related to the task and selects the optimal generative AI algorithm. The selection unit can also refer to patent documents related to the task and select the optimal generative AI algorithm. Furthermore, the selection unit can also refer to technical documents related to the task and select the optimal generative AI algorithm. As a result, the system can improve the accuracy of the selection by referring to the related literature of the task and select the optimal generative AI algorithm.

[0088] The acquisition unit can estimate the user's emotion and adjust the timing of acquiring the execution result in consideration of the estimated user's emotion. The estimation of emotion is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is stressed, the timing of acquiring the execution result is delayed to wait until the user is relaxed. Also, when the user is relaxed, the execution result is acquired immediately, and the task is processed efficiently. Furthermore, when the user is in a hurry, the execution result is acquired immediately, and the adjustment is made to obtain the result as quickly as possible. As a result, the system can acquire the execution result at a more appropriate timing by adjusting the timing of acquiring the execution result according to the user's emotion.

[0089] The acquisition unit can select the optimal acquisition method by referring to the past execution results of each generating AI algorithm at the time of acquisition. The past execution results include, for example, log data and history data. As a result, the acquisition unit can select the optimal acquisition method by referring to the past execution results, thereby enabling efficient acquisition. For example, the acquisition unit analyzes the past execution time of each generating AI algorithm and selects the most efficient acquisition method. The acquisition unit can also select the most reliable acquisition method by referring to the past accuracy of each generating AI algorithm. Furthermore, the acquisition unit can also select the most balanced acquisition method by taking into account the past resource usage rate of each generating AI algorithm. As a result, the system can select the optimal acquisition method by referring to the past execution results, thereby enabling efficient acquisition.

[0090] The acquisition unit can apply different acquisition algorithms depending on the type of task at the time of acquisition. The acquisition algorithms include, for example, scraping and data mining. As a result, the acquisition unit can perform efficient acquisition by applying the optimal acquisition algorithm depending on the type of task. For example, the acquisition unit applies a high-speed acquisition algorithm to a task with a large amount of calculation. In addition, the acquisition unit can also apply an acquisition algorithm with high communication efficiency to a task with a large amount of data transfer. Furthermore, the acquisition unit can also apply a low-latency acquisition algorithm to a task that requires real-time performance. As a result, the system can perform efficient acquisition by applying the optimal acquisition algorithm depending on the type of task.

[0091] The acquisition unit can estimate the user's emotion and adjust the acquisition order of the execution results in consideration of the estimated user's emotion. The estimation of emotion is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is stressed, the execution results are acquired in order from the simplest task. Also, when the user is relaxed, the execution results are acquired in order from the most complicated task. Furthermore, when the user is in a hurry, the execution results are acquired in order from the most time-consuming task. In this way, the system can acquire the execution results in a more appropriate order by adjusting the acquisition order of the execution results according to the user's emotion.

[0092] The acquisition unit can acquire the execution results by taking into consideration the geographical distribution of the tasks. The geographical distribution includes, for example, communication delays in each region and the location of the data center. As a result, the acquisition unit can acquire the execution results with the communication delays minimized by taking into consideration the geographical distribution of the tasks. For example, the acquisition unit preferentially acquires the execution results of the generation AI algorithms that are geographically close, thereby minimizing the communication delays. The acquisition unit can also acquire the execution results of the generation AI algorithms that are geographically distributed, thereby distributing the load. Furthermore, the acquisition unit can acquire the execution results of the generation AI algorithms that are located in geographically important locations, thereby prioritizing processing in a specific region. As a result, the system can acquire the execution results with the communication delays minimized by taking into consideration the geographical distribution of the tasks.

[0093] The acquisition unit can improve the accuracy of acquisition by referring to the related literature of the task at the time of acquisition. The related literature includes, for example, academic papers, technical reports, patent documents, etc. As a result, the acquisition unit can improve the accuracy of acquisition and select the optimal acquisition method by referring to the related literature of the task. For example, the acquisition unit refers to the latest research paper related to the task and selects the optimal acquisition method. The acquisition unit can also refer to patent documents related to the task and select the optimal acquisition method. Furthermore, the acquisition unit can also refer to technical documents related to the task and select the optimal acquisition method. As a result, the system can improve the accuracy of acquisition and select the optimal acquisition method by referring to the related literature of the task.

[0094] The optimization unit can estimate the user's emotion and adjust the optimization criteria in consideration of the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is stressed, an optimization criterion that minimizes resource usage is applied. Also, when the user is relaxed, an optimization criterion that emphasizes accuracy is applied. Furthermore, when the user is in a hurry, an optimization criterion that emphasizes speed is applied. As a result, the system adjusts the optimization criterion according to the user's emotion, thereby performing optimization with a more appropriate criterion.

[0095] During optimization, the optimization unit can optimize the optimization algorithm by referring to past execution result data. The past execution result data includes, for example, log data and history data. As a result, the optimization unit can improve the accuracy of the optimization algorithm by referring to the past execution result data, enabling efficient optimization. For example, the optimization unit analyzes the past execution result data and selects the most efficient optimization algorithm. The optimization unit can also select an optimization algorithm with low resource usage from the past execution result data. Furthermore, the optimization unit can select an optimization algorithm with a low error rate based on the past execution result data. As a result, the system can improve the accuracy of the optimization algorithm by referring to the past execution result data, enabling efficient optimization.

[0096] The optimization unit can apply different optimization algorithms depending on the type of task during optimization. Examples of optimization algorithms include genetic algorithms and simulated annealing. As a result, the optimization unit can apply an optimal optimization algorithm depending on the type of task, thereby enabling efficient optimization. For example, the optimization unit applies a high-speed optimization algorithm to a task with a large amount of calculations. In addition, the optimization unit can also apply an optimization algorithm with high communication efficiency to a task with a large amount of data transfer. Furthermore, the optimization unit can also apply a low-latency optimization algorithm to a task that requires real-time performance. As a result, the system can apply an optimal optimization algorithm depending on the type of task, thereby enabling efficient optimization.

[0097] The optimization unit can estimate the user's emotions and determine the priority of optimization in consideration of the estimated user's emotions. The estimation of emotions is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is stressed, a task with low importance is optimized preferentially to reduce the burden on the user. In addition, when the user is relaxed, a task with high importance is optimized preferentially to efficiently complete the task. Furthermore, when the user is in a hurry, the most time-consuming task is optimized preferentially to shorten the overall processing time. As a result, the system can optimize tasks with more appropriate priorities by determining the priority of optimization according to the user's emotions.

[0098] The optimization unit can perform optimization by taking into account the geographical distribution of tasks. The geographical distribution includes, for example, communication delays by region and the location of data centers. As a result, the optimization unit can perform optimization with minimized communication delays by taking into account the geographical distribution of tasks. For example, the optimization unit prioritizes optimization of geographically close tasks to minimize communication delays. The optimization unit can also optimize geographically distributed tasks to distribute the load. Furthermore, the optimization unit can optimize tasks located in geographically important locations to prioritize processing in specific regions. As a result, the system can perform optimization with minimized communication delays by taking into account the geographical distribution of tasks.

[0099] The optimization unit can improve the accuracy of optimization by referring to the related literature of the task during optimization. The related literature includes, for example, academic papers, technical reports, patent documents, and the like. As a result, the optimization unit can improve the accuracy of optimization and enable efficient optimization by referring to the related literature of the task. For example, the optimization unit can select an optimal optimization method by referring to the latest research papers related to the task. The optimization unit can also select an optimal optimization method by referring to patent documents related to the task. Furthermore, the optimization unit can also select an optimal optimization method by referring to technical documents related to the task. As a result, the system can improve the accuracy of optimization and enable efficient optimization by referring to the related literature of the task.

[0100] The interface can estimate the user's emotion and adjust the display method of the interface taking into account the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is nervous, an interface with calm colors is provided to reduce visual stress. Also, when the user is having fun, an interface with bright colors is provided to make input work fun. Furthermore, when the user is tired, a simple and highly visible interface is provided to make input work easier. As a result, the system can display more appropriately by adjusting the display method of the interface according to the user's emotion.

[0101] When displaying the interface, the interface can select the optimal display method by referring to the user's past operation history. The operation history includes, for example, a click history and an operation log. This allows the interface to select the optimal display method by referring to the past operation history, enabling efficient display. For example, the interface preferentially provides a display method that the user has frequently used in the past. The interface can also suggest the most efficient display method from the user's past operation history. Furthermore, the interface can analyze the user's past operation history and select a display method with high visibility. This allows the system to select the optimal display method by referring to the past operation history, enabling efficient display.

[0102] The interface can estimate the user's emotions and adjust the operation procedure of the interface taking into account the estimated user's emotions. The estimation of emotions is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. For example, when the user is nervous, a simple and intuitive operation procedure is provided. Also, when the user is enjoying himself, detailed operation procedures are provided to make the operation enjoyable. Furthermore, when the user is tired, a minimum operation procedure is provided to reduce the burden of operation. As a result, the system adjusts the operation procedure of the interface according to the user's emotions, enabling more appropriate operation.

[0103] The interface can select an optimal display method by considering device information of the user when displaying the interface. The device information includes, for example, the type of device, the screen size, and the OS version. As a result, the interface can select an optimal display method by considering the device information of the user, and can provide efficient display. For example, when the user is using a smartphone, the interface can provide a display method that matches the screen size. In addition, when the user is using a tablet, the interface can also provide a display method that is optimized for a large screen. Furthermore, when the user is using a smart watch, the interface can also provide a simple and highly visible display method. As a result, the system can select an optimal display method by considering the device information of the user, and can provide efficient display.

[0104] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0105] The system can estimate the user's emotions and dynamically change the priority of tasks based on the estimated emotions. For example, if the user is stressed, the system can prioritize simple tasks to reduce stress. Also, if the user is relaxed, the system can prioritize complex tasks. Furthermore, if the user is in a hurry, the system can prioritize the most time-consuming tasks to reduce the overall processing time. In this way, the system dynamically changes the priority of tasks according to the user's emotions, enabling more appropriate task processing.

[0106] The system can monitor the user's emotions in real time while the task is being performed and adjust the task progress according to changes in emotions. For example, if the user begins to feel stressed while the task is in progress, the system can temporarily suspend the task and suggest a break to relax. Also, if the user is relaxed, the task progress can be accelerated. Furthermore, if the user is in a hurry, the system can optimize the task progress and adjust it to complete it in the shortest time. This allows the system to adjust the task progress according to the user's emotions, enabling more effective task management.

[0107] The system can estimate the user's emotions and dynamically change task allocation based on the estimated emotions. For example, if the user is stressed, the system can assign easy tasks to other users to reduce stress. Also, if the user is relaxed, the system can assign complex tasks to the user. Furthermore, if the user is in a hurry, the system can assign the most time-consuming tasks to other users to reduce the overall processing time. In this way, the system dynamically changes task allocation according to the user's emotions, enabling more appropriate task management.

[0108] The system can estimate the user's emotions and provide an interface that visualizes the task progress based on the estimated emotions. For example, if the user is stressed, the system can display the progress in a simple and intuitive graphic to reduce visual stress. If the user is relaxed, the system can display detailed progress so that the user can enjoy the task progress. Furthermore, if the user is in a hurry, the system can update the progress in real time to allow the user to quickly grasp the situation. This allows the system to adjust the visualization of the progress according to the user's emotions and provide a more appropriate interface.

[0109] The system can estimate the user's emotions and adjust the task notification method based on the estimated emotions. For example, if the user is feeling stressed, the system can reduce notifications to reduce the user's burden. Also, if the user is relaxed, the system can proactively notify the user and promote task progress. Furthermore, if the user is in a hurry, the system can prioritize important notifications and encourage a quick response. This allows the system to adjust the notification method according to the user's emotions, enabling more appropriate task management.

[0110] The system can monitor resource usage in real time while a task is being executed and optimally allocate resources. For example, if a particular generative AI algorithm is consuming excessive resources, the system will reallocate resources to other algorithms. Also, if resource usage is low, the system can effectively utilize resources by executing additional tasks. Furthermore, depending on resource usage, the system can dynamically change task priorities to achieve efficient task processing. This allows the system to optimally allocate resources and enable efficient task management.

[0111] The system can detect errors in real time while a task is being performed and take measures to minimize the impact of the error. For example, if a particular generative AI algorithm generates an error, the system can switch to another algorithm. Also, if the frequency of errors is high, the system can identify the cause of the error and provide feedback to correct it. Furthermore, to minimize the impact of errors, the system can automatically rerun the task. This allows the system to minimize the impact of errors and enable efficient task processing.

[0112] The system can check the consistency of data in real time while a task is being executed, and automatically correct any inconsistencies that occur. For example, if a data inconsistency is detected, the system automatically corrects the data and continues executing the task. In addition, if data inconsistencies occur frequently, the system can identify the cause of the inconsistency and provide feedback to correct it. Furthermore, to maintain data consistency, the system can periodically check the data and make corrections as necessary. This allows the system to maintain data consistency and enable efficient task processing.

[0113] The system can monitor performance in real time while a task is being executed and optimize performance. For example, if the performance of a particular generative AI algorithm degrades, the system will switch to another algorithm. Also, if performance degradation occurs frequently, the system can identify the cause of the performance degradation and provide feedback to correct it. Furthermore, to optimize performance, the system can periodically check performance and optimize as necessary. This allows the system to optimize performance and enable efficient task processing.

[0114] The system can detect security risks in real time while a task is being performed and take measures to minimize the impact of the risk. For example, if a specific generative AI algorithm generates a security risk, the system can switch to another algorithm. Also, if the frequency of security risks is high, the system can identify the cause of the risk and provide feedback to correct it. Furthermore, to minimize the impact of security risks, the system can automatically rerun the task. This allows the system to minimize the impact of security risks and enable efficient task processing.

[0115] The process flow of the second embodiment will be briefly described below.

[0116] Step 1: The parallel processing unit executes multiple generative AI algorithms simultaneously for a task with a time limit. For example, the parallel processing unit can execute the generative AI algorithms simultaneously using multiple processors. Step 2: The selection unit selects a generation AI algorithm from among the multiple generation AI algorithms executed by the parallel processing unit based on the execution speed. For example, the selection unit can select an optimal generation AI algorithm based on the execution speed of each generation AI algorithm as well as accuracy and resource usage. Step 3: The acquisition unit acquires the execution result of the task by the generation AI algorithm selected by the selection unit. For example, the acquisition unit can monitor the execution result of the selected generation AI algorithm in real time and perform reselection as necessary.

[0117] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0118] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) are input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may also be an AI agent. In addition, when the processes of each of the above-mentioned parts are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples.In addition, processing performed by AI including the generation AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI including the generation AI.

[0119] Furthermore, the above-described processing by the data processing system 10 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information required for processing from the data processing device 12 or an external device.

[0120] For example, the parallel processing unit is realized by either the smart device 14 or the data processing device 12. For example, a plurality of generative AI algorithms can be executed simultaneously using the processor 46 of the smart device 14 or the processor 28 of the data processing device 12. The selection unit is realized by, for example, the control unit 46A of the smart device 14 or the specific processing unit 290 of the data processing device 12, and selects the optimal generative AI algorithm based on the execution speed. The acquisition unit is realized by, for example, the control unit 46A of the smart device 14 or the specific processing unit 290 of the data processing device 12, and acquires the execution result of the selected generative AI algorithm. The optimization unit is realized by, for example, the specific processing unit 290 of the data processing device 12, and learns past execution result data, and selects the generative AI algorithm and optimizes the execution order. The interface visualizes the execution result of the generative AI algorithm using, for example, the display 40A of the smart device 14 or the display of the data processing device 12. The parallel processing unit can, for example, estimate the user's emotion and adjust the start timing of the parallel processing based on the estimated emotion. The emotion estimation is performed, for example, using the camera 42 or the microphone 38B of the smart device 14. The interface can, for example, estimate the user's emotion and adjust the display method based on the estimated emotion. The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0121] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0122] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0123] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0124] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0125] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0126] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0127] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0128] Fig. 4 shows an example of main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0129] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0130] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0131] In the smart glasses 214, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart glasses 214 have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0132] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0133] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0134] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0135] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device.

[0136] For example, the parallel processing unit is realized by either the smart glasses 214 or the data processing device 12. For example, a plurality of generative AI algorithms can be executed simultaneously using the processor 46 of the smart glasses 214 or the processor 28 of the data processing device 12. The selection unit is realized by, for example, the control unit 46A of the smart glasses 214 or the specific processing unit 290 of the data processing device 12, and selects an optimal generative AI algorithm based on the execution speed. The acquisition unit is realized by, for example, the control unit 46A of the smart glasses 214 or the specific processing unit 290 of the data processing device 12, and acquires the execution result of the selected generative AI algorithm. The optimization unit is realized by, for example, the specific processing unit 290 of the data processing device 12, and learns past execution result data, and selects the generative AI algorithm and optimizes the execution order. The interface visualizes the execution result of the generative AI algorithm using, for example, the display of the smart glasses 214 or the display of the data processing device 12. The parallel processing unit can, for example, estimate the user's emotion and adjust the start timing of the parallel processing based on the estimated emotion. The emotion estimation is performed, for example, using the camera 42 and microphone 238 of the smart glasses 214. The interface can, for example, estimate the user's emotion and adjust the display method based on the estimated emotion. The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0137] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0138] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0139] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0140] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0141] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0142] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0143] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0144] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0145] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0146] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0147] In the headset type terminal 314, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50, and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0148] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0149] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0150] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0151] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device.

[0152] For example, the parallel processing unit is realized by either the headset type terminal 314 or the data processing device 12. For example, a plurality of generative AI algorithms can be executed simultaneously using the processor 46 of the headset type terminal 314 or the processor 28 of the data processing device 12. The selection unit is realized by, for example, the control unit 46A of the headset type terminal 314 or the specific processing unit 290 of the data processing device 12, and selects the optimal generative AI algorithm based on the execution speed. The acquisition unit is realized by, for example, the control unit 46A of the headset type terminal 314 or the specific processing unit 290 of the data processing device 12, and acquires the execution result of the selected generative AI algorithm. The optimization unit is realized by, for example, the specific processing unit 290 of the data processing device 12, and learns past execution result data, and selects the generative AI algorithm and optimizes the execution order. The interface visualizes the execution result of the generative AI algorithm using, for example, the display 343 of the headset type terminal 314 or the display of the data processing device 12. The parallel processing unit can, for example, estimate the user's emotion and adjust the start timing of the parallel processing based on the estimated emotion. The emotion estimation is performed, for example, using the camera 42 and microphone 238 of the headset terminal 314. The interface can, for example, estimate the user's emotion and adjust the display method based on the estimated emotion. The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0153] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0154] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0155] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0156] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0157] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0158] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of a typical healthy person).

[0159] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0160] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and legs. The posture and behavior of the robot 414 are controlled by controlling the motors of the arms, hands, and legs. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0161] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0162] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0163] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0164] In the robot 414, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads out the specific program 60 from the storage 50, and executes the specific program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific program 60 executed on the RAM 48. The robot 414 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0165] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0166] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0167] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0168] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0169] For example, the parallel processing unit is realized by either the robot 414 or the data processing device 12. For example, a plurality of generative AI algorithms can be executed simultaneously using the processor 46 of the robot 414 or the processor 28 of the data processing device 12. The selection unit is realized by, for example, the control unit 46A of the robot 414 or the specific processing unit 290 of the data processing device 12, and selects the optimal generative AI algorithm based on the execution speed. The acquisition unit is realized by, for example, the control unit 46A of the robot 414 or the specific processing unit 290 of the data processing device 12, and acquires the execution result of the selected generative AI algorithm. The optimization unit is realized by, for example, the specific processing unit 290 of the data processing device 12, and learns past execution result data, and selects the generative AI algorithm and optimizes the execution order. The interface visualizes the execution result of the generative AI algorithm using, for example, the display of the robot 414 or the display of the data processing device 12. The parallel processing unit can, for example, estimate the user's emotion and adjust the start timing of the parallel processing based on the estimated emotion. The emotion estimation is performed using, for example, the camera 42 or the microphone 238 of the robot 414. The interface can, for example, estimate the user's emotion and adjust the display method based on the estimated emotion. The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0170] The emotion identification model 59 as an emotion engine may determine the emotion of the user according to a specific mapping. Specifically, the emotion identification model 59 may determine the emotion of the user according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the emotion of the robot, and the identification processing unit 290 may perform identification processing using the emotion of the robot.

[0171] FIG. 9 is a diagram showing an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive emotions are arranged. The more outside the concentric circles, the more emotions that represent states and actions that arise from a state of mind are arranged. Emotions are a concept that includes emotions and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions that occur in the brain are arranged. On the right side of the concentric circles, emotions that are generally induced by situational judgment are arranged. On the upper and lower sides of the concentric circles, emotions that are generally generated from reactions that occur in the brain and are induced by situational judgment are arranged. In addition, on the upper side of the concentric circles, emotions of "pleasure" are arranged, and on the lower side, emotions of "discomfort" are arranged. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0172] These emotions are distributed in the 3 o'clock direction of emotion map 400 and usually fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0173] The inside of emotion map 400 represents what is going on inside one's mind, and the outside of emotion map 400 represents behavior, so the further out you go on emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0174] Here, human emotions are based on various balances such as posture and blood sugar level, and when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. Emotions can also be created for robots, cars, motorcycles, etc., based on various balances such as posture and battery level, so that when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. The emotion map may be generated, for example, based on the emotion map of Dr. Mitsuyoshi (Research on speech emotion recognition and emotion brain physiological signal analysis system, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). On the left half of the emotion map, emotions belonging to an area called "reaction" where sensation is dominant are lined up. On the right half of the emotion map, emotions belonging to an area called "situation" where situation recognition is dominant are lined up.

[0175] The emotion map defines two emotions that promote learning. The first is the negative emotion around the middle of "repentance" or "remorse" on the situation side. In other words, this is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the positive emotion around "desire" on the response side. In other words, this is when the robot has positive feelings such as "I want more" or "I want to know more."

[0176] The emotion identification model 59 inputs the user input to a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the emotion of the user. This neural network is pre-trained based on multiple learning data that are combinations of the user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in Fig. 10. Fig. 10 shows an example in which multiple emotions, "relief," "calm," and "encouraging," have similar emotion values.

[0177] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers, including computer 22.

[0178] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable computer-readable non-transitory storage medium such as a Universal Serial Bus (USB) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0179] In addition, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 upon request from the data processing device 12.

[0180] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0181] As the hardware resource for executing the specific process, various processors as shown below can be used. An example of the processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing the specific process by executing software, i.e., a program. Another example of the processor is a dedicated electric circuit, which is a processor having a circuit configuration designed exclusively for executing a specific process, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), or an Application Specific Integrated Circuit (ASIC). Each processor has a built-in or connected memory, and each processor executes the specific process by using the memory.

[0182] The hardware resource that executes the specific process may be one of these various processors, or may be a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0183] As an example of a configuration using one processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a configuration in which a processor is used that realizes the functions of the entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0184] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements. The specific processes described above are merely examples. It goes without saying that unnecessary steps may be deleted, new steps may be added, or the order of processes may be changed without departing from the spirit of the invention.

[0185] In the above example, the first to fourth embodiments are described separately, but some or all of these embodiments may be combined. The smart device 14, the smart glasses 214, the headset terminal 314, and the robot 414 are only examples, and they may be combined with each other, or may be other devices. In the above example, the first and second embodiments are described separately, but they may be combined.

[0186] The above description and illustrations are detailed descriptions of the parts related to the technology of the present disclosure, and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, function, action, and effect is an example of the configuration, function, action, and effect of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replaced with respect to the description and illustrations shown above, within the scope of the gist of the technology of the present disclosure. In addition, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the description and illustrations shown above omit explanations of technical common sense that do not require particular explanation in enabling the implementation of the technology of the present disclosure.

[0187] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and standard was specifically and individually indicated to be incorporated by reference.

[0188] (Appendix 1) For tasks with time limits, A parallel processing unit for simultaneously executing multiple generative AI algorithms; a selection unit that selects a generation AI algorithm from among the plurality of generation AI algorithms executed by the parallel processing unit based on an execution speed; an acquisition unit that acquires an execution result of the task by the generation AI algorithm selected by the selection unit; A system characterized by: (Appendix 2) Considering the results of execution of the plurality of generative AI algorithms for a plurality of tasks, An optimization unit that optimizes the selection of a generation AI algorithm for each of the plurality of tasks and the order of execution of each of the plurality of tasks. 2. The system of claim 1, (Appendix 3) The parallel processing unit includes: Use multiple processors to run the generative AI algorithm simultaneously 2. The system of claim 1, (Appendix 4) The selection unit is Select a generative AI algorithm by considering its execution speed, as well as its accuracy and resource usage. 2. The system of claim 1, (Appendix 5) The selection unit is Monitor the execution results of the selected generative AI algorithm in real time and reselect it if necessary. 2. The system of claim 1, (Appendix 6) The optimization unit is Includes a machine learning model that learns from past execution result data and optimizes the selection of generative AI algorithms and execution order. 3. The system of claim 2, (Appendix 7) Provides an interface for visualizing the results of generative AI algorithms 2. The system of claim 1, (Appendix 8) The parallel processing unit includes: The system estimates the user's emotions and adjusts the start timing of parallel processing based on the estimated user emotions. 2. The system of claim 1, (Appendix 9) The parallel processing unit includes: During parallel processing, the optimal processor allocation is performed by referring to the past execution results of each generative AI algorithm. 2. The system of claim 1, (Appendix 10) The parallel processing unit includes: When parallel processing, different parallel processing algorithms are applied depending on the type of task. 2. The system of claim 1, (Appendix 11) The parallel processing unit includes: Estimate the user's emotions and determine the priority of parallel processing based on the estimated user's emotions. 2. The system of claim 1, (Appendix 12) The parallel processing unit includes: During parallel processing, processor allocation is performed taking into account the geographic distribution of tasks. 2. The system of claim 1, (Appendix 13) The parallel processing unit includes: When performing parallel processing, select the optimal parallel processing method by referring to the relevant literature for the task. 2. The system of claim 1, (Appendix 14) The selection unit is Estimate user emotions and adjust the selection criteria of the generative AI algorithm taking into account the estimated user emotions. 2. The system of claim 1, (Appendix 15) The selection unit is When making a selection, the past results of each generative AI algorithm are referenced to improve the accuracy of the selection. 2. The system of claim 1, (Appendix 16) The selection unit is During selection, apply different selection algorithms depending on the type of task 2. The system of claim 1, (Appendix 17) The selection unit is Estimate user emotions and adjust the selection order of generative AI algorithms based on the estimated user emotions. 2. The system of claim 1, (Appendix 18) The selection unit is When selecting, the generative AI algorithm will be chosen taking into account the geographical distribution of tasks. 2. The system of claim 1, (Appendix 19) The selection unit is When making a selection, refer to the relevant literature for the task to improve the accuracy of the selection. 2. The system of claim 1, (Appendix 20) The acquisition unit is Estimate the user's emotions and adjust the timing of obtaining the execution results based on the estimated user emotions. 2. The system of claim 1, (Appendix 21) The acquisition unit is When acquiring data, the optimal acquisition method is selected by referring to the past execution results of each generation AI algorithm. 2. The system of claim 1, (Appendix 22) The acquisition unit is During acquisition, different acquisition algorithms are applied depending on the type of task. 2. The system of claim 1, (Appendix 23) The acquisition unit is The user's emotions are estimated, and the order in which execution results are obtained is adjusted based on the estimated user's emotions. 2. The system of claim 1, (Appendix 24) The acquisition unit is When obtaining, the geographic distribution of tasks is taken into account to obtain the execution results. 2. The system of claim 1, (Appendix 25) The acquisition unit is During acquisition, refer to relevant literature for the task to improve the accuracy of acquisition. 2. The system of claim 1, (Appendix 26) The optimization unit is Infer user emotions and adjust optimization criteria based on the estimated user emotions 3. The system of claim 2, (Appendix 27) The optimization unit is During optimization, the optimization algorithm is optimized by referring to past execution result data. 3. The system of claim 2, (Appendix 28) The optimization unit is During optimization, different optimization algorithms are applied depending on the type of task. 3. The system of claim 2, (Appendix 29) The optimization unit is Estimate user emotions and determine optimization priorities based on the estimated user emotions 3. The system of claim 2, (Appendix 30) The optimization unit is During optimization, the geographic distribution of tasks is taken into account 3. The system of claim 2, (Appendix 31) The optimization unit is During optimization, refer to the relevant literature for the task to improve the accuracy of the optimization 3. The system of claim 2, (Appendix 32) The interface is The system estimates the user's emotions and adjusts the interface display method based on the estimated user's emotions. 2. The system of claim 1, (Appendix 33) The interface is When displaying an interface, the optimal display method is selected by referring to the user's past operation history. 2. The system of claim 1, (Appendix 34) The interface is Estimating user emotions and adjusting interface operation procedures based on the estimated user emotions 2. The system of claim 1, (Appendix 35) The interface is When displaying the interface, the optimal display method is selected taking into account the user's device information. 2. The system of claim 1, [Explanation of symbols]

[0189] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. For tasks with time limits, A parallel processing unit for simultaneously executing a plurality of generating AI algorithms; a selection unit that selects a generation AI algorithm from among the plurality of generation AI algorithms executed by the parallel processing unit based on an execution speed; An acquisition unit that acquires an execution result of the task by the generation AI algorithm selected by the selection unit. A system characterized by:

2. An optimization unit that optimizes the selection of a generation AI algorithm for each of the plurality of tasks and the order of execution of each of the plurality of tasks based on the execution results of the plurality of generation AI algorithms for the plurality of tasks.

2. The system of claim 1 .

3. The parallel processing unit includes: Using multiple processors to run the generative AI algorithm simultaneously 2. The system of claim 1 .

4. The selection unit is Select the generation AI algorithm taking into consideration the execution speed of each generation AI algorithm as well as accuracy and resource utilization.

2. The system of claim 1 .

5. The selection unit is Monitor the execution results of the selected generative AI algorithm in real time and reselect it if necessary.

2. The system of claim 1 .

6. The optimization unit is Includes a machine learning model that learns from past execution result data and optimizes the selection of generating AI algorithms and execution order.

3. The system of claim 2.

7. Provides an interface for visualizing the results of generative AI algorithm execution 2. The system of claim 1 .

8. The parallel processing unit includes: The system estimates the user's emotions and adjusts the start timing of parallel processing based on the estimated user emotions.

2. The system of claim 1 .

9. The parallel processing unit includes: During parallel processing, the optimal processor allocation is performed by referring to the past execution results of each generation AI algorithm.

2. The system of claim 1 .

10. The parallel processing unit includes: When parallel processing, different parallel processing algorithms are applied depending on the type of task.

2. The system of claim 1 .

Images