Electronic device and control method of electronic device

Abstract

An electronic device and a method for controlling the electronic device are provided. The electronic device includes, based on first device information regarding the hardware specifications of the electronic device, second device information regarding the hardware specifications of the first external device, first model information regarding one or more neural network models included in the first external device, and second model information regarding one or more neural network models included in the electronic device, identifying whether each of the one or more neural network models included in the first external device is suitable for the hardware of the electronic device, and whether each of the one or more neural network models identified as suitable for the hardware of the electronic device is suitable for replacing the neural network included in the electronic device.

Description

Technical Field

[0001] This invention relates to an electronic device and a control method thereof. More specifically, this invention relates to an electronic device and a control method thereof for identifying the transformation suitability of a neural network model included in an external device. Background Technology

[0002] In recent years, there has been a growing demand from users and industries for technologies that provide customized services to each user through personalized neural network models. However, on the other hand, the need to protect personal data related to user privacy is also increasing. Therefore, in many cases, the focus is limited to training neural network models by collecting the personal data required for personalization.

[0003] Therefore, research is underway to develop a technology that enables neural network models to be used on devices without transferring personal data to external servers or the cloud, and a technology that transmits information from a specific device to another device via device-to-device communication.

[0004] When transmitting information about a personalized neural network model from one device to another, it may be unsuitable to transmit the model if the devices are of different types, or even if they are of the same type but have different hardware specifications. Therefore, it is necessary to perform an identification process to determine whether the neural network model is suitable for transmission between different devices when transmitting information about a personalized neural network model from one device to another.

[0005] Furthermore, when users of electronic devices purchase new electronic devices, there is a need for a technology that can efficiently and reliably transfer personalized neural network models from existing electronic devices to new ones.

[0006] The above information is provided for background purposes only to aid in understanding this disclosure. No decision has been made, nor any assertion, is made regarding whether any of the foregoing items can be considered as prior art applicable to this disclosure. Summary of the Invention

[0007] [Technical Issues]

[0008] The present invention aims to at least address the aforementioned problems and / or disadvantages, and to provide at least the following advantages. Therefore, one aspect of the present invention is to provide an electronic device and a control method thereof, which is capable of identifying, based on information about the hardware specifications of each electronic device and external device, and information about a neural network model, whether it is appropriate to transmit a neural network model included in an external device to the electronic device.

[0009] Additional aspects will be set forth in part in the description which follows, and in part will be obvious from the description or may be learned by practice of the presented embodiments.

[0010] [Technical Solutions]

[0011] According to one aspect of the present invention, an electronic device is provided for identifying the transformation suitability of a neural network model included in an external device. The electronic device includes: a communicator; a memory configured to store first device information regarding the hardware specifications of the electronic device and a hardware suitability identifier for identifying neural network models suitable for the hardware of the electronic device; and a processor configured to, based on received user input, control the communicator to send a first signal requesting information related to one or more neural network models included in one or more external devices; in response to the first signal, receive a second signal from a first external device among the one or more external devices via the communicator, the signal including second device information regarding the hardware specifications of the first external device and first model information regarding one or more neural network models included in the first external device; by inputting the first device information, the second device information, and the first model information into the hardware suitability identifier, identify whether each of the one or more neural network models included in the first external device is suitable for the hardware of the electronic device; and control the communicator to send a signal to the first external device including information related to the transformation suitability of the neural network models included in the first external device. A third signal is received from a first external device via a communicator requesting installation data of one or more neural network models identified as suitable for hardware in an electronic device, and in response to the third signal, a fourth signal is received from the first external device including installation data of one or more neural network models identified as suitable for hardware in an electronic device. The processor is configured to identify one or more neural network models included in the first external device as suitable for hardware in an electronic device based on the specification of each of a plurality of hardware configurations included in the electronic device being greater than or equal to the specification of the plurality of hardware configurations included in the first external device; and to identify one or more neural network models included in the first external device whose hardware requirement specifications are lower than the specifications of the plurality of hardware configurations included in the first external device as suitable for hardware in an electronic device based on the specification of one or more of the plurality of hardware configurations included in the electronic device being lower than the specifications of the plurality of hardware configurations included in the first external device.

[0012] According to another aspect of the present invention, a control method for an electronic device is provided, the electronic device storing first device information regarding the hardware specifications of the electronic device and a hardware suitability identifier for identifying neural network models suitable for the hardware of the electronic device, and identifying the transformation suitability of neural network models included in external devices. The control method includes sending a first signal based on received user input for requesting information related to one or more neural network models included in one or more external devices; in response to the first signal, receiving a second signal from a first external device, including second device information regarding the hardware specifications of the first external device and first model information regarding one or more neural network models included in the first external device; identifying whether each of the one or more neural network models included in the first external device is suitable for the hardware of the electronic device by inputting the first device information, the second device information, and the first model information into the hardware suitability identifier; sending a third signal to the first external device including a request for installation data of one or more neural network models identified as suitable for the hardware of the electronic device; and in response to the third signal, receiving information from the first external device... The device receives a fourth signal including installation data of one or more neural network models identified as suitable for hardware in the electronic device. Identifying whether each of the one or more neural network models is suitable for the hardware in the electronic device includes: identifying one or more neural network models included in the first external device as suitable for hardware in the electronic device based on the specifications of each of the plurality of hardware configurations included in the electronic device being greater than or equal to the specifications of the plurality of hardware configurations included in the first external device; and identifying one or more neural network models included in the first external device whose hardware requirement specifications are lower than the specifications of the plurality of hardware configurations included in the first external device as suitable for hardware in the electronic device based on the specifications of one or more of the plurality of hardware configurations included in the electronic device being lower than the specifications of the plurality of hardware configurations included in the first external device.

[0013] According to another aspect of this disclosure, a non-transitory computer-readable recording medium includes a program for performing a control method for an electronic device, the electronic device storing first device information regarding the hardware specifications of the electronic device and a hardware suitability identifier for identifying neural network models suitable for the hardware of the electronic device, and identifying the transformation suitability of neural network models included in external devices. The control method includes sending a first signal based on received user input for requesting information related to one or more neural network models including one or more external devices; in response to the first signal, receiving a second signal from a first external device including second device information regarding the hardware specifications of the first external device and first model information regarding one or more neural network models included in the first external device; identifying whether each of the one or more neural network models included in the first external device is suitable for the hardware of the electronic device by inputting the first device information, the second device information, and the first model information into the hardware suitability identifier; sending a third signal to the first external device including a request for installation data of one or more neural network models identified as suitable for the hardware of the electronic device; and in response to the third signal, receiving information from the first external device... An external device receives a fourth signal including installation data of one or more neural network models identified as suitable for hardware of an electronic device, wherein identifying whether each of the one or more neural network models is suitable for hardware of the electronic device includes identifying one or more neural network models included in the first external device as suitable for hardware of the electronic device based on the specification of each of a plurality of hardware configurations included in the electronic device being greater than or equal to the specification of a plurality of hardware configurations included in the first external device, and identifying one or more neural network models included in the first external device whose hardware requirement specifications are lower than the specifications of a plurality of hardware configurations included in the first external device as suitable for hardware of the electronic device.

[0014] According to another aspect of the invention, an electronic device is provided for identifying the conversion suitability of a neural network model included in an external device. The electronic device includes a communicator; a memory configured to store internal model information about one or more neural network models included in the electronic device and a model suitability identifier for identifying whether a neural network model is suitable for replacing a neural network model included in the electronic device; and a processor configured to, based on received user input, control the communicator to send a first signal for requesting information related to one or more neural network models included in one or more external devices; in response to the first signal, receive a second signal via the communicator including external model information about one or more neural network models included in the first external device; identify whether each of the one or more neural network models included in the first external device is suitable for replacing a neural network model included in the electronic device by inputting the internal model information and the external model information into the model suitability identifier; and control the communicator to send a third signal including a request for installation data identified as suitable for replacing one or more neural network models included in the electronic device. The processor is sent to a first external device, and in response to a third signal, receives a fourth signal from the first external device via a communicator, which includes installation data of one or more identified neural network models. The processor is configured to compare the service type of one or more neural network models included in the electronic device with the service type of one or more neural network models included in the first external device based on service type information included in each of the internal model information and the external model information; to compare the personalization level of the first neural network model with the personalization level of the second neural network model based on the fact that the service type of the first neural network model included in the one or more neural network models in the first external device is the same as the service type of the second neural network model included in the plurality of neural network models in the electronic device; to compare the personalization level of the first neural network model with the personalization level of the second neural network model based on information about personalization level included in each of the internal model information and the external model information; and to identify the first neural network model as suitable to replace the second neural network model based on the fact that the personalization level of the first neural network model is higher than the personalization level of the second neural network model.

[0015] Other aspects, advantages and distinctive features of the invention will become apparent to those skilled in the art from the following detailed description, which discloses various embodiments of the invention in conjunction with the accompanying drawings. Attached Figure Description

[0016] The above and other aspects, features, and advantages of certain embodiments of the present invention will become more apparent from the following description taken in conjunction with the accompanying drawings, wherein:

[0017] Figure 1This is a flowchart illustrating a control method for an electronic device according to an embodiment of the present invention;

[0018] Figure 2 , 3 4 and 5 are for specifically describing the various embodiments of the present invention for performing according to... Figure 1 A diagram illustrating the operation of a hardware suitability identification module in the control method of an electronic device;

[0019] Figure 5A is a flowchart illustrating a control method for an electronic device according to an embodiment of the present invention;

[0020] Figure 5B is a flowchart illustrating a control method for an electronic device according to an embodiment of the present invention;

[0021] Figure 6 , 7 Figures 5A and 5B are used to specifically describe the operation of a model suitability identification module for performing a control method for an electronic device according to various embodiments of the present invention;

[0022] Figure 9 This is a flowchart describing a control method for an electronic device according to an embodiment of the present invention;

[0023] Figure 10 This is a sequence diagram illustrating an example of a situation where multiple external devices exist according to embodiments of the present invention;

[0024] Figure 11 This is a diagram used to describe a user interface provided by an electronic device according to an embodiment of the present invention;

[0025] Figure 12 This is a diagram illustrating a user interface provided by a first external device according to an embodiment of the present invention;

[0026] Figure 13 This is a diagram illustrating an example of an electronic device, a first external device, and a second external device according to an embodiment of the present invention;

[0027] Figure 14 It is a sequence diagram used to describe the process by which an electronic device identifies the conversion suitability when a neural network model included in a first external device is transmitted to a second external device according to an embodiment of the present invention;

[0028] Figure 15 This is a block diagram illustrating in detail the architecture of the software modules included in an electronic device according to an embodiment of the present invention;

[0029] Figure 16 This is a block diagram illustrating in detail the architecture of a software module included in a first external device according to an embodiment of the present invention;

[0030] Figure 17 This is a block diagram schematically illustrating the architecture of a hardware configuration included in an electronic device according to an embodiment of the present invention; and

[0031] Figure 18 This is a block diagram illustrating in more detail the architecture of the hardware configuration included in an electronic device according to an embodiment of the present invention.

[0032] Throughout the accompanying drawings, similar reference numerals will be understood to refer to similar parts, components, and structures. Detailed Implementation

[0033] The following description, with reference to the accompanying drawings, helps to provide a comprehensive understanding of the various embodiments of the invention as defined by the claims and their equivalents. It includes various specific details to aid understanding, but these are merely examples. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the various embodiments described herein without departing from the scope and spirit of the invention. Furthermore, for clarity and brevity, descriptions of known functions and constructions may be omitted.

[0034] The terms and words used in the following description and claims are not limited to their meanings in the references. The inventors use these terms and words only to provide a clear and consistent understanding of the invention. Therefore, those skilled in the art should understand that the following description of various embodiments of the invention is for illustrative purposes only and is not intended to limit the invention as defined in the appended claims and their equivalents.

[0035] It should be understood that the singular forms “a,” “one,” and “the” include plural references unless the context explicitly specifies otherwise. Thus, for example, a reference to “component surface” includes a reference to one or more such surfaces.

[0036] The terminology used in this invention is for describing particular embodiments only and is not intended to be limiting. Singular expressions include plural expressions unless the context clearly indicates otherwise.

[0037] In this invention, the expressions "having", "may have", "including", "may include", etc., indicate the presence of corresponding features (e.g., numerical values, functions, operations, components such as parts, etc.) and do not exclude the presence of additional features.

[0038] In this invention, expressions such as “A or B”, “at least one of A and / or B”, “one or more of A and / or B” can include all possible combinations of the items listed together. For example, “A or B”, “at least one of A and B” or “at least one of A or B” can refer to all cases (1) including at least one A, (2) including at least one B, or (3) including at least one A and at least one B.

[0039] The terms “first”, “second”, etc. used in this invention can refer to various components without regard to the order and / or importance of the components. They are only used to distinguish one component from other components and do not limit the corresponding components.

[0040] When reference is made to any component (e.g., the first component) being coupled or connected (operationally or communicatively) to another component (e.g., the second component), it should be understood that any component is directly coupled / connected to the other component, or may be coupled / connected to the other component through another component (e.g., the third component).

[0041] On the other hand, when it is mentioned that any component (e.g., the first component) is "directly coupled" or "directly connected" to another component (e.g., the second component), it should be understood that the other component (e.g., the third component) does not exist between any component and the other component.

[0042] Depending on the specific circumstances, the expression "configured (or set) as" used in this invention may be replaced with "suitable for," "capable of," "designed to," "suitable for," "manufactured as," or "able to." The term "configured (or set) as" does not necessarily refer only to "specifically designed for" in hardware.

[0043] Alternatively, in any context, the phrase “configured to” may mean that the device is “capable” of working with other devices or components. For example, “processor configured (or set) to perform A, B, and C” could refer to a dedicated processor (e.g., an embedded processor) used to perform the respective operations, or a general-purpose processor (e.g., a central processing unit (CPU) or application processor) that can perform the respective operations by executing one or more software stored in a memory device.

[0044] In the embodiments, a "module" or "device" may perform at least one function or operation and may be implemented as hardware or software, or as a combination of hardware and software. Furthermore, multiple "modules" or multiple "devices" may be integrated into at least one module and may be implemented as at least one processor, except for "modules" or "devices" that require implementation as specific hardware.

[0045] On the other hand, the various elements and regions in the accompanying drawings are shown in schematic form. Therefore, the technical spirit of the invention is not limited to the relative sizes or spacing shown in the accompanying drawings.

[0046] Electronic devices according to different embodiments of the present invention may include at least one of, for example, a smartphone, a tablet computer (PC), a desktop computer, a laptop computer, or a wearable device. Wearable devices may include at least one of accessory wearable devices (e.g., watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted devices (HMDs)), textile or clothing-integrated wearable devices (e.g., electronic clothing), body-attached wearable devices (e.g., skin pads or tattoos), or bio-implanted circuitry.

[0047] In some embodiments, electronic devices may include a television (TV), a digital video disc (DVD) player, an audio player, a refrigerator, an air conditioner, a cleaner, an oven, a microwave oven, a washing machine, an air filter, a set-top box, a home automation control panel, a security control panel, a media box (such as Samsung Electronics Ltd.'s HomeSync™, Apple Inc.'s TV™ or Google™'s TV™), a game console (such as Xbox™ and PlayStation™), an electronic dictionary, an electronic key, a camera, or a digital photo frame.

[0048] In other embodiments, the electronic device may include at least one of various medical devices (e.g., various portable medical measurement devices such as blood glucose meters, heart rate monitors, blood pressure monitors, thermometers, etc.), magnetic resonance angiography (MRA), magnetic resonance imaging (MRI), computed tomography (CT), imaging equipment, ultrasound equipment, etc.), navigation devices, global navigation satellite systems (GNSS), event data loggers (EDR), flight data loggers (FDR), automotive infotainment devices, marine electronic devices (e.g., marine navigation devices, gyrocompasses, etc.), avionics, security devices, automotive mainframes, industrial or household robots, drones, automated teller machines (ATMs) of financial institutions, point-of-sale (POS) terminals of stores, or Internet of Things (IoT) devices (e.g., light bulbs, various sensors, sprinkler systems, fire alarms, thermostats, streetlights, toasters, sports equipment, hot water tanks, heaters, boilers, etc.).

[0049] In the following description, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can readily implement the present invention.

[0050] Figure 1 This is a flowchart illustrating a control method for an electronic device according to an embodiment of the present invention.

[0051] Figures 2 to 4 This is for specifically describing the various embodiments of the present invention for performing according to... Figure 1 The control method of the electronic device 100 (see Figure 17 Hardware suitability identification module 1100 (see) Figure 2The diagram shows the operation of ).

[0052] First, the "electronic device" according to the present invention can be implemented in various types, such as smartphones, tablets, laptops, televisions, and robots, and is not limited to a specific type of device. In the following text, the electronic device according to the present invention is referred to as electronic device 100.

[0053] A neural network model refers to an artificial intelligence model, including artificial neural networks, and can be trained through deep learning. For example, a neural network model may include at least one of the following artificial neural network models: deep neural network (DNN), convolutional neural network (CNN), recurrent neural network (RNN), or generative adversarial network (GAN). However, the neural network model according to the present invention is not limited to the examples described above.

[0054] In the electronic device 100 according to the present invention, first device information regarding the hardware specifications of the electronic device 100 and the hardware suitability identification module 1100 may be stored.

[0055] "First device information" refers to information about the hardware specifications of electronic device 100. Specifically, first device information is a term used to refer generally to information about specifications indicating the performance of each of the plurality of hardware components included in electronic device 100, and may include detailed information about the presence, quantity, type, and performance of each of the plurality of hardware components included in electronic device 100.

[0056] Specifically, the first device information may include specifications of the processor included in the electronic device 100, specifications of the memory included in the electronic device 100, and specifications of the data acquisition device included in the electronic device 100. The data acquisition device is a component that acquires data input to one or more neural network models included in the electronic device 100, and may include at least one of a camera, microphone, or sensor included in the electronic device 100.

[0057] The first device information may include performance evaluation information about each of the multiple hardware components included in the electronic device 100. This "performance evaluation information" is a score obtained through a comprehensive evaluation of the performance of each hardware configuration based on expert experiments and analysis, and may be pre-stored in the electronic device 100 and received and updated from an external server.

[0058] "Hardware suitability identification module 1100" refers to a module that identifies a neural network model suitable for the hardware of electronic device 100. Specifically, "hardware suitability identification module 1100" can output information about whether it is suitable for executing a neural network model included in an external device by using the hardware of electronic device 100. In describing the present invention, the term "suitable" may be replaced by terms such as "compatible" or "alternative".

[0059] refer to Figure 1 In operation S110, when user input is received, electronic device 100 may send a first signal to request information related to one or more neural network models included in one or more external devices.

[0060] "User input" can be received based on user touch input via the display of electronic device 100, user voice received via the microphone of electronic device 100, input via physical buttons provided in electronic device 100, control signals sent by a remote control device for controlling electronic device 100, etc. The term "transmission" can be used to mean both unicast (where signals or data are transmitted for a specific external device) and broadcast (where signals or data are transmitted simultaneously to all external devices connected to the network without targeting a specific external device). "Information related to one or more neural network models" can include second device information and first model information, which will be described later.

[0061] Similar to electronic device 100, the "external device" can be of various types, such as smartphones, tablets, laptops, televisions, and robots, and the type of external device may also differ from that of electronic device 100. Electronic device 100 and the external device can be "connected" to each other, meaning that a communication connection is established by exchanging requests and responses for a communication connection between electronic device 100 and the external device. The communication connection method according to the invention is not particularly limited.

[0062] In response to the first signal, during operation S120, the electronic device 100 can receive a second signal from a first external device among one or more external devices. This second signal includes second device information regarding the hardware specifications of the first external device and first model information regarding one or more neural network models included in the first external device. That is, if a first signal corresponding to a request to search for one or more neural network models is received, the first external device can send a second signal to the electronic device 100 as a response to the request, and the second signal may include the second device information and the first model information. In describing the invention, the term "first external device" is used to specify an external device capable of transmitting the installation data of a neural network model to the electronic device 100. Hereinafter, the first external device according to the invention will be referred to as first external device 200-1 (see [link to original text]). Figure 14 ).

[0063] The first external device 200-1 can perform a user authentication process based on first user information included in the first signal and second user information stored in the first external device 200-1, and can also send the second signal to the electronic device 100 when user authentication is completed. (Refer to...) Figure 9 and 10 Describe in detail the user authentication process or user suitability identification process.

[0064] "Second device information" refers to information about the hardware specifications of the first external device 200-1. Specifically, second device information is a term used to refer generally to information about specifications indicating the performance characteristics of each piece of hardware included in the first external device 200-1, and corresponds to the first device information about the hardware performance of the electronic device 100.

[0065] In other words, the second device information may include detailed information about the presence, quantity, type, and performance of each of the plurality of hardware configurations included in the first external device 200-1, and similar to the first device information, the second device information may also include performance evaluation information for each piece of hardware included in the first external device 200-1. Furthermore, the second device information may include information about the specifications of the processor included in the first external device 200-1, the specifications of the memory included in the first external device 200-1, and the specifications of the data acquisition device included in the first external device 200-1. The data acquisition device of the first external device 200-1 is a component that acquires data input to one or more neural network models included in the first external device 200-1, similar to the data acquisition device of the electronic device 100, and may include at least one of a camera, microphone, or sensor included in the first external device 200-1.

[0066] "First model information" refers to information about one or more neural network models included in the first external device 200-1. Specifically, the first model information may include information about the type of service, information about the level of personalization, and information about the hardware requirements specifications of each of the one or more neural network models included in the first external device 200-1.

[0067] At operation S130, if a second signal including second device information and first model information is received from the first external device 200-1, the electronic device 100 can input the first device information, the second device information, and the first model information into the hardware suitability identification module 1100 to identify whether each of the one or more neural network models included in the first external device 200-1 is suitable for the hardware of the electronic device 100. That is, if the first device information, the second device information, and the first model information are input, the hardware suitability identification module 1100 can perform a hardware suitability identification process according to different embodiments of this disclosure based on the first device information, the second device information, and the first model information. Hereinafter, reference will be made to... Figures 2 to 4 as well as Figure 1 The "hardware suitability identification process" is described in detail. Specifically, the hardware suitability identification process according to the present invention may include, for example: Figure 2 and Figure 3 Operations 1 and 2 are shown.

[0068] First, refer to Figure 1 and 2 Describe "Step 1". By identifying whether the hardware specifications of the electronic device 100 are equal to or better than the hardware specifications of the first external device 200-1 in all parts based on the first device information stored in the electronic device 100 and the second device information received from the first external device 200-1, the electronic device 100 can perform a hardware suitability identification process for all neural network models included in the first external device 200-1.

[0069] Specifically, if the specification of each of the plurality of hardware configurations included in the electronic device 100 is greater than or equal to the specification of the plurality of hardware configurations included in the first external device 200-1 (Y in S140), at operation S150-1, the electronic device 100 may identify one or more neural network models included in the first external device 200-1 as hardware suitable for the electronic device 100.

[0070] In other words, electronic device 100 can identify the hardware configuration of a first external device 200-1 corresponding to each of the plurality of hardware configurations included in electronic device 100, and can compare the specifications of each corresponding hardware configuration. Furthermore, as a result of comparing the specifications of each of the plurality of hardware configurations included in electronic device 100 with the specifications of each of the plurality of hardware configurations included in first external device 200-1, if the specifications of each of the plurality of hardware configurations included in electronic device 100 are greater than or equal to the specifications of the plurality of hardware configurations included in first external device 200-1, this can be estimated to include a situation where all or more neural network models in first external device 200-1 can be executed using the hardware of electronic device 100, regardless of the hardware specifications required for each of the one or more neural network models in first external device 200-1. Therefore, in this case, electronic device 100 can identify all or more neural network models in first external device 200-1 as suitable for the hardware of electronic device 100.

[0071] refer to Figure 4 The hardware suitability identification module 1100 may include a processor suitability identification module 1110, a memory suitability identification module 1120, a camera suitability identification module 1130, a microphone suitability identification module 1140, and a sensor suitability identification module 1150. Furthermore, the electronic device 100 can perform a hardware suitability identification process for each hardware configuration based on information about processor specifications, memory specifications, camera specifications, microphone specifications, and sensor specifications included in the first and second device information, through each module included in the hardware suitability identification module 1100.

[0072] For example, if the specifications of the processor, memory, camera, microphone, and sensor included in electronic device 100 are all greater than or equal to the specifications of the processor, memory, camera, microphone, and sensor included in the first external device 200-1, electronic device 100 can identify all or more neural network models included in the first external device 200-1 as hardware suitable for electronic device 100. In this case, electronic device 100 can perform a hardware suitability identification process for each hardware configuration in the order of processor, memory, camera, microphone, and sensor; however, for each hardware configuration according to the present invention, there is no specific order restriction in the hardware suitability identification process.

[0073] The foregoing has described a comparison of the specifications of the processor, memory, camera, microphone, and sensor included in each of the electronic device 100 and the first external device 200-1, but this is only for the sake of description, and according to the present invention, the electronic device 100 can compare the specifications of each detailed component including the processor, memory, camera, microphone, and sensor.

[0074] For example, electronic device 100 can also perform a hardware suitability identification process by comparing the specifications of electronic device 100 and the first external device 200-1 for the central processing unit (CPU), graphics processing unit (GPU), and neural processing unit (NPU) in the processor; comparing the specifications of electronic device 100 and the first external device 200-1 for the random access memory (RAM) and read-only memory (ROM) in the memory; and comparing the specifications of electronic device 100 and the first external device 200-1 for each of the sensors (such as a global positioning system (GPS) sensor, a gyroscope sensor, an accelerometer sensor, and a lidar sensor). Examples of detailed components included in the processor, memory, camera, microphone, and sensors are not limited to the examples described above.

[0075] According to one embodiment of the present invention, the specifications of the hardware configuration included in the electronic device 100 are greater than or equal to the specifications of the hardware configuration included in the first external device 200-1. This may mean that all specifications indicating the performance of the hardware configuration included in the electronic device 100 are superior to or at least equal to all specifications indicating the performance of the hardware configuration included in the first external device 200-1.

[0076] For example, if the CPU included in electronic device 100 has 8 cores, 16 threads, a clock speed of 3.6 GHz, and a cache capacity of 8 MB, and the CPU included in the first external device 200-1 has 4 cores, 8 threads, a clock speed of 3.3 GHz, and a cache capacity of 8 MB, electronic device 100 can determine that the performance of the CPU in electronic device 100 is greater than or equal to the performance of the CPU in the first external device 200-1. In this example, the number of cores, the number of threads, the clock speed, and the cache are used as examples of CPU performance metrics, but other performance metrics such as bus speed and thermal design power (TDP) can also be considered.

[0077] According to another embodiment, the specifications of the hardware configuration included in the electronic device 100 are greater than or equal to the specifications of the hardware configuration included in the first external device 200-1. This may mean that performance evaluation information indicating the performance of the hardware configuration included in the electronic device 100 is higher than performance evaluation information indicating the performance of the hardware configuration included in the first external device 200-1. As described above, the performance evaluation information is a score obtained by comprehensively evaluating the performance of each hardware configuration based on expert experiments and analysis, and may be included in the first device information and the second device information. For example, if the CPU included in the electronic device 100 has a score of 97 according to the performance evaluation information, and the CPU included in the first external device 200-1 has a score of 86 according to the performance evaluation information, the electronic device 100 can identify that the performance of the CPU of the electronic device 100 is higher than the performance of the CPU of the first external device 200-1.

[0078] Secondly, according to an embodiment of the present invention, reference will be made to Figure 1 and 3 Describe "Operation 2". By identifying whether the hardware specifications of the electronic device 100 are equal to or better than the hardware requirement specifications of each of one or more neural network models included in the first external device 200-1 based on the first device information stored in the electronic device 100 and the first model information received from the first external device 200-1, the electronic device 100 can perform a hardware suitability identification process for each neural network model included in the first external device 200-1.

[0079] refer to Figure 3 Step 2 of the hardware suitability identification process can only be executed if step 1 of the hardware suitability identification process described above fails. That is, as a result of executing step 1 of the hardware suitability identification process described above, if it is identified that at least one neural network model among one or more neural network models included in the first external device 200-1 is not suitable for the hardware of the electronic device 100, the electronic device 100 can execute step 2 of the hardware suitability identification process.

[0080] Specifically, if at least one specification of the plurality of hardware configurations included in the electronic device 100 is smaller than the specification of the plurality of hardware configurations included in the first external device 200-1 (N in S140), at operation S150-2, the electronic device 100 may identify at least one neural network model whose hardware requirement specification is lower than the specification of the plurality of hardware configurations included in the first external device 200-1 as hardware suitable for the electronic device 100.

[0081] In other words, as a result of comparing the specifications of each of the multiple hardware configurations included in the electronic device 100 with the specifications of each of the multiple hardware configurations included in the first external device 200-1, if the specifications of at least one of the multiple hardware configurations included in the electronic device 100 are smaller than the specifications of the multiple hardware configurations included in the first external device 200-1, this can be said to be a case in which all neural network models suitable for execution using the hardware of the electronic device 100 can be identified only when the hardware requirement specifications of each of the one or more neural network models included in the first external device 200-1 are individually compared with the specifications of the multiple hardware configurations included in the electronic device 100.

[0082] Therefore, in this case, the electronic device 100 can identify whether all the hardware requirements of the one or more neural network models included in the first external device 200-1 are less than the specifications of the multiple hardware configurations included in the electronic device 100 by comparing the hardware requirement specifications of the one or more neural network models included in the first external device 200-1 with the specifications of the multiple hardware configurations included in the electronic device 100.

[0083] For example, if all the hardware requirements of the first neural network model included in the first external device 200-1 are smaller than the specifications of the plurality of hardware configurations included in the electronic device 100, the electronic device 100 may identify the first neural network model as hardware suitable for the electronic device 100. On the other hand, if at least one of the hardware requirements of the second neural network model included in the first external device 200-1 is greater than or equal to the specifications of the plurality of hardware configurations included in the electronic device 100, the electronic device 100 may identify the second neural network model as hardware unsuitable for the electronic device 100. As described above, information regarding the hardware requirements of each of the one or more neural network models included in the first external device 200-1 may be included in the first model information received from the first external device 200-1.

[0084] The various methods for comparing the specifications of the multiple hardware configurations included in the electronic device 100 with the specifications of each of the multiple hardware configurations included in the first external device 200-1 in step 1 can be applied to the method for comparing the specifications of the multiple hardware configurations included in the electronic device 100 with the hardware requirement specifications of one or more neural network models included in the first external device 200-1 in step 2. Therefore, redundant descriptions of the method for comparing the specifications of the multiple hardware configurations included in the electronic device 100 with the hardware requirement specifications of one or more neural network models included in the first external device 200-1 in step 2 are omitted.

[0085] When comparing the hardware requirements specifications of each of the one or more neural network models included in the first external device 200-1 with the specifications of the multiple hardware configurations included in the electronic device 100, the hardware specifications of the electronic device 100 at the time of product release can be compared with the hardware requirements specifications of each of the one or more neural network models included in the first external device 200-1. However, when comparing the hardware requirements specifications of each of the one or more neural network models included in the first external device 200-1 with the specifications of the multiple hardware configurations included in the electronic device 100, the hardware specifications of the electronic device 100 can also be compared with the hardware requirements specifications of each of the one or more neural network models included in the first external device 200-1.

[0086] As described above, the comparison of the specifications of the multiple hardware configurations included in the electronic device 100 with the specifications of the multiple hardware configurations included in the first external device 200-1 in step 1 is based on the premise that the configurations correspond to each other and that there is a comparison object. That is, if there is a hardware configuration included in the first external device 200-1 but not included in the electronic device 100, the electronic device 100 can perform the hardware suitability identification process according to step 2, unless the hardware configuration can be replaced by another hardware configuration included in the electronic device 100.

[0087] Furthermore, as described above, the comparison in step 2 of the specifications of the multiple hardware configurations included in the electronic device 100 with the hardware requirement specifications of one or more neural network models included in the first external device 200-1 is also based on the premise that the configurations correspond to each other and that there is a comparison object. That is, if there is a hardware configuration requested in the hardware requirement specification of the neural network model of the first external device 200-1 but not included in the electronic device 100, the electronic device 100 may identify the neural network model as hardware unsuitable for the electronic device 100, unless the hardware configuration can be replaced by another hardware configuration included in the electronic device 100.

[0088] As described above, if at least one neural network model suitable for the hardware of electronic device 100 is identified among one or more neural network models included in the first external device 200-1, at operation S160, electronic device 100 may send a third signal to the first external device 200-1 including a request for installation data of the identified one or more neural network models. Furthermore, in response to the third signal, at operation S170, electronic device 100 may receive a fourth signal from the first external device 200-1 including installation data of at least one neural network model identified as suitable for the hardware of electronic device 100.

[0089] In other words, if a third signal including a request for installation data of one or more identified neural network models is sent to the first external device 200-1, the first external device 200-1 can send the installation data of one or more identified neural network models to the electronic device 100. Furthermore, if installation data of one or more identified neural network models is received, the electronic device 100 can install one or more identified neural network models based on the received installation data. Therefore, the electronic device 100 can execute at least one neural network model personalized by the first external device 200-1 in the electronic device 100 by replacing one or more of the plurality of neural network models included in the electronic device 100 with one or more identified neural network models. According to embodiments of the present invention, the personalization of neural network models will be described later with reference to Figures 5A to 8.

[0090] Specifically, the installation data received from the first external device 200-1 may include identification information for each of one or more identified neural network models, as well as configuration information required to install one or more identified neural network models. The "configuration information" may include information about the structure and type of the neural network included in the neural network model, the number of layers included in the neural network, the number of nodes per layer, the weight value of each node, and the connection relationships between multiple nodes.

[0091] Simultaneously, "installing" a neural network model may include "switching" the neural network model included in electronic device 100 to the neural network model included in the first external device 200-1. Furthermore, "switching" the neural network model included in electronic device 100 to the neural network model included in the external device may include replacing some neural network models of electronic device 100 with some neural network models of the external device, such as changing only the weight values ​​of nodes in the neural network model included in electronic device 100 to the weight values ​​of nodes in the neural network model included in the external device, and replacing the entire neural network model of electronic device 100 with the entire neural network model of the external device.

[0092] According to the above reference Figures 1 to 4 In the various embodiments described, when transmitting a neural network model from an external device, the electronic device 100 can determine whether to transmit the neural network model by identifying whether it is suitable to execute the neural network model using the hardware of the electronic device 100, and thus, the efficiency and reliability of transmitting the learning may be significantly improved.

[0093] Figure 5A is a flowchart illustrating a control method for an electronic device according to an embodiment of the present invention.

[0094] Figure 5B is a flowchart illustrating a control method for an electronic device according to an embodiment of the present invention.

[0095] Figures 6 to 8 This is a diagram used to specifically describe the operation of a model suitability identification module for performing a control method for an electronic device according to FIG5A, according to various embodiments of the present invention.

[0096] In the above text, refer to Figures 1 to 4 The hardware suitability identification process according to the present invention has been described. However, according to another embodiment of the present invention, the electronic device 100 may perform a "model suitability identification process" for each neural network model included in a first external device of the electronic device 100. Hereinafter, the model suitability identification process according to the present invention will be described in detail with reference to Figures 5A to 8.

[0097] First, the electronic device 100 can also store the second model information and the model fitness identification module 1200, as well as the first device information, the second device information and the hardware fitness identification module 1100 as described above.

[0098] "Second model information" refers to information about one or more neural network models included in the electronic device 100. Specifically, the second model information may include information about the type of service and information about the level of personalization of each of the one or more neural network models included in the electronic device 100.

[0099] "Model fitness identification module 1200" refers to the module that identifies a neural network model suitable for replacing the neural network model included in the electronic device 100. (Reference) Figure 6 Based on the second model information stored in the electronic device 100 and the first model information received from the first external device 200-1, the "model suitability identification module 1200" can output information about whether it is suitable to replace the neural network model included in the electronic device 100 with the neural network model included in the external device.

[0100] Referring to Figure 5A, if one or more neural network models (U in S210) suitable for the hardware of electronic device 100 are identified, at operation S220, electronic device 100 can input first model information and second model information into model suitability identification module 1200 to identify whether each of the one or more neural network models identified as suitable for the hardware of electronic device 100 is suitable for replacing the neural network model included in electronic device 100.

[0101] Specifically, such as Figure 7As shown, the model fitness identification process according to the present invention may include two processes executed by each of the model type fitness identification module 1210 and the personalization level fitness identification module 1220. Therefore, in the following, after describing each process included in the model fitness identification process in detail, the control method according to an embodiment of the present invention will be described again.

[0102] First, such as Figure 7 As shown, based on the information about service types included in each of the first model information and the second model information, the "model type suitability identification module 1210" can identify neural network models with the same service type by comparing the service type of each of one or more neural network models included in the electronic device 100 with the service type of each of one or more neural network models identified as suitable for the hardware of the electronic device 100.

[0103] The "type of neural network model" can be classified based on the input / output information and the corresponding function of the neural network model. For example, when both the first neural network model and the second neural network model are neural network models that provide speech recognition services by inputting speech signals based on the user's speech and outputting text corresponding to the user's speech, the model type suitability identification module 1210 can identify that the service type of the first neural network model and the service type of the second neural network model are the same.

[0104] Secondly, such as Figure 7 As shown, based on the information about the personalization level included in each of the first model information and the second model information, the "personalization level suitability identification module 1220" can identify a neural network model with a higher personalization level among neural network models with the same service type by comparing the personalization levels between neural network models with the same service type.

[0105] "Personalization level of the neural network model" refers to the degree to which the neural network model is updated based on the user when training the neural network model using personalized data related to the user of the electronic device 100 after the neural network model is installed in the electronic device 100. "Personalized data" refers to the training data used to personalize the neural network model included in the external device. According to embodiments of the present invention, reference will be made to... Figure 7 and 8 The operation of the personalization level suitability identification module 1220 is described in more detail.

[0106] Specifically, the level of personalization can be determined based on "information about the user's history." Information about the user's history can include the number of times, frequency, and duration of use of the neural network model the user employs. For example, if a second neural network model included in electronic device 100 is used 100 times in one month, it can be determined that this represents a higher level of personalization than a first neural network model included in an external device that is used 30 times in one month or 100 times in two months.

[0107] refer to Figure 8 Information regarding the history of users using the neural network model 2000 included in the electronic device 100 may include instances where the user used the neural network model 2000 based on data obtained through another electronic device 100 connected to the electronic device 100. Figure 8 Type 2 in the above), and the case where the user uses the neural network model 2000 included in the electronic device 100 based on data obtained through the electronic device 100 ( Figure 8 Type 1 in the middle).

[0108] For example, the use of the neural network model 2000 included in the smartphone when an image acquired by a user through a robotic cleaner connected to the smartphone is transmitted to the smartphone, and the use of the neural network model 2000 included in the smartphone based on an image acquired through the smartphone, can be included in the history of using the neural network model 2000 included in the smartphone.

[0109] The level of personalization can also be determined based on "information about user feedback." This information can include direct evaluation information of the user's usage results after using the neural network model 2000, as well as indirect evaluation information related to the usage results. Figure 8 Type 3 in the middle).

[0110] For example, when the neural network model 2000 is used in the electronic device 100, if a user directly inputs positive evaluation information about an application that includes the neural network model 2000, the personalization level of the neural network model 2000 may increase. On the other hand, when the neural network model 2000 is used in the electronic device 100, if a user directly inputs negative evaluation information about an application that includes the neural network model 2000, the personalization level of the neural network model 2000 may increase. Conversely, when the same user input is repeatedly input into the neural network model 2000, it can be indirectly confirmed that the user is dissatisfied with the output of the neural network model 2000, and therefore, the personalization level of the neural network model 2000 may decrease.

[0111] The level of personalization can be quantitatively calculated based on a weighted sum of various types of information used to assess the aforementioned level of personalization. Specifically, refer to... Figure 8 The level of personalization can be calculated by summing the values ​​obtained by multiplying each "number of times per type" and the preset "weight of each type".

[0112] For example, the highest weight can be assigned when a user inputs direct evaluation information for an application including the neural network model 2000; the second highest weight can be assigned when data acquired from the electronic device 100 (such as images captured using the camera of the electronic device 100 or speech acquired using the microphone of the electronic device 100) is input into the neural network model 2000; and the third highest weight can be assigned when data received from another electronic device 100 (not the electronic device 100) connected to the electronic device 100 is input into the neural network model 2000. Furthermore, the value obtained by multiplying the number of each type by the weight of each type can be a weighted sum of each type, and the value obtained by summing the weighted sums of all types can be a quantitative value indicating the personalization level of the neural network model 2000.

[0113] Referring again to Figure 5A, if the service type of a first neural network model among one or more neural network models identified as suitable for the hardware of electronic device 100 is the same as the service type of a second neural network model among a plurality of neural network models included in electronic device 100 (Y in operation S230), electronic device 100 can compare the personalization level of the first neural network model and the personalization level of the second neural network model based on information about personalization level included in each of the first model information and the second model information. Furthermore, if the personalization level of the first neural network model is higher than the personalization level of the second neural network model (Y in operation S240), at operation S250, it can be identified that the first neural network model is suitable to replace the second neural network model.

[0114] At operation S260, if the above-described model suitability identification process has been performed, the electronic device 100 may send a third signal to the first external device 200-1. This third signal includes a request for installation data of at least one neural network model identified as suitable for replacing the neural network model included in the electronic device 100. That is, when the third signal is sent to the first external device 200-1 after performing not only the hardware suitability identification process but also the model suitability identification process, the third signal may include a request for installation data of at least one neural network model identified during the model suitability identification process.

[0115] In response to the third signal, at operation S270, the electronic device 100 can receive a fourth signal from the first external device 200-1, which includes installation data of one or more identified neural network models. That is, when the third signal is sent to the first external device 200-1 after performing not only the hardware suitability identification process but also the model suitability identification process, the fourth signal received in response to the third signal may include installation data of at least one neural network model identified during the model suitability identification process.

[0116] The embodiments of performing a model suitability identification process after the hardware suitability identification process have been described above. However, according to the present invention, there is no time sequence limitation between the hardware suitability identification process and the model suitability identification process. Specifically, the electronic device 100 may first input the first model information and the second model information into the model suitability identification module 1200 to identify whether it is suitable to replace the first neural network model with the second neural network model. Furthermore, if it is determined that it is suitable to replace the first neural network model with the second neural network model, the electronic device 100 may also determine whether it is suitable to use the hardware of the electronic device 100 to execute the second neural network model by comparing the hardware performance of the electronic device 100 with the hardware performance required to execute the second neural network model.

[0117] As described above, the personalization level of the first neural network model and the second neural network model included in the first external device 200-1 are compared only when the service types of the first neural network model and the second neural network model included in the electronic device 100 are the same. However, according to another embodiment of the present invention, when there is no neural network model in the electronic device 100 having the same service type as the first neural network model included in the first external device 200-1, the electronic device 100 can also recognize that it is appropriate to transmit the first neural network model to the electronic device 100 without recognizing the personalization level of the first neural network model.

[0118] In the foregoing, embodiments have been described in which, after performing the hardware suitability identification process, an additional model suitability process is performed for each of one or more neural network models identified as suitable for the hardware of electronic device 100, but this disclosure is not limited thereto. That is, according to another embodiment of the invention, electronic device 100 may also receive installation data of a neural network model that satisfies model suitability from first external device 200-1 by performing only the model suitability identification process without performing the hardware suitability identification process. Hereinafter, an embodiment of receiving installation data from first external device 200-1 by performing only the model suitability identification process will be described with reference to FIG. 5B. However, in the following description of FIG. 5B, redundant descriptions of the same content as described above will be omitted.

[0119] Referring to FIG5B, at operation S205, when user input is received, electronic device 100 may send a first signal for requesting information related to one or more neural network models included in one or more external devices. Furthermore, at operation S215, in response to the first signal, electronic device 100 may receive a second signal including external model information regarding one or more neural network models included in the first external device 200-1 from one or more external devices. Hereinafter, within the limitations used to describe the embodiment shown in FIG5B, the term "external model information" has the same meaning as... Figure 1 The term "first model information" in the description up to 5A has the same meaning.

[0120] At operation S220, if a second signal including external model information is received from the first external device 200-1, the electronic device 100 can input the internal model information stored in the electronic device and the external model information received from the first external device 200-1 into the model suitability identification module 1200 to identify whether each of one or more neural network models identified as suitable for the hardware of the model electronic device 100 is suitable to replace the neural network model included in the electronic device 100. Hereinafter, within the limitations of the embodiment shown in FIG5B, the term "internal model information" is used with the same meaning as the term "second model information" in the description of FIG5A.

[0121] As described above, the model fitness identification process according to the present invention may include a model type fitness identification process and a personalization level fitness identification process executed by each of the model type fitness identification module 1210 and the personalization level fitness identification module 1220, such as Figure 7 As shown.

[0122] Specifically, if the service type of a first neural network model included in one or more neural network models in the first external device 200-1 is the same as the service type of a second neural network model included in multiple neural network models in the electronic device 100 (Y in operation S230), the electronic device 100 can compare the personalization level of the first neural network model and the personalization level of the second neural network model based on information about personalization level included in each of the internal model information and the external model information. Furthermore, at operation S250, if the personalization level of the first neural network model is higher than the personalization level of the second neural network model (Y in operation S240), it can be identified that the first neural network model is suitable to replace the second neural network model.

[0123] At operation S260, if the above-described model suitability identification process has been performed, the electronic device 100 may send a third signal to the first external device 200-1, the third signal including a request for installation data of at least one neural network model identified as suitable for replacing the neural network model included in the electronic device 100. Furthermore, at operation S270, in response to the third signal, the electronic device 100 may receive a fourth signal from the first external device 200-1 including installation data of one or more identified neural network models.

[0124] According to the various embodiments described above with reference to Figures 5A to 8, when transmitting a neural network model from an external device, the electronic device 100 can determine whether to transmit the neural network model by identifying whether it is suitable to replace the neural network model included in the electronic device 100 based on the degree of personalization of the neural network model included in the external device, and accordingly, the efficiency and reliability of transmitting the learning may be further improved.

[0125] Figure 9 This is a flowchart describing a control method for an electronic device 100 according to an embodiment of the present invention.

[0126] Figure 10 This is a sequence diagram illustrating an example of a situation where multiple external devices exist according to embodiments of the present invention.

[0127] The hardware suitability identification process and model suitability identification process according to the present invention have been described above. However, according to another embodiment of the present invention, a user suitability identification process may also be performed before the hardware suitability identification process and the model suitability identification process. The "user suitability identification process" is the process of identifying whether a user of the electronic device 100 is suitable for using the neural network model of the external device, and may be referred to as the "user authentication process." In this invention, the term "conversion compatibility" can be used to encompass the meanings of hardware suitability, model suitability, and user suitability. Reference will be made below to... Figure 9 and 10 Describe the user suitability identification process in detail.

[0128] exist Figures 1 to 8 The description is based on the premise that when electronic device 100 sends a first signal to first external device 200-1 to request information related to one or more neural network models included in first external device 200-1, the external device sends a second signal to electronic device 100 in response to the first signal, including second device information about the hardware specifications of first external device 200-1 and first model information about one or more neural network models included in first external device 200-1.

[0129] However, before the external device provides the second device information and the first model information to the electronic device 100, the user suitability identification process can be performed by the external device, or the user suitability identification process can be performed by the electronic device 100. During the user suitability identification process, since both the first user information and the second user information can be used, the first user information may include at least one of account information about the user of the electronic device 100 or identification information about the electronic external device, and the second user information may include at least one of account information about the user of the first external device 200-1 and identification information about the first external device 200-1.

[0130] refer to Figure 9 The electronic device 100 according to the present invention can perform a user suitability identification process. Specifically, as Figure 9 As shown, the user suitability identification process can be performed by the user suitability identification module 1300 included in the electronic device 100.

[0131] "User suitability identification module 1300" refers to a module that identifies whether a user of electronic device 100 is suitable for using the neural network model included in the first external device 200-1, based on first user information stored in electronic device 100 and second user information received from first external device 200-1.

[0132] Specifically, the first user information can be stored in the electronic device 100. Furthermore, the electronic device 100 can send a first signal to the first external device 200-1 requesting information related to one or more neural network models included in the first external device 200-1. If the first signal is received, the first external device 200-1 can send second user information to the electronic device 100. If the second user information is received, the electronic device 100 can perform a user suitability identification process based on the first and second user information.

[0133] Specifically, if the account information of the user of electronic device 100 included in the first user information matches the account information of the user of the first external device 200-1 included in the second user information, electronic device 100 can identify that the user of electronic device 100 is suitable for using the neural network model included in the first external device 200-1. On the other hand, if the account information of the user of electronic device 100 included in the first user information does not match the account information of the user of the first external device 200-1 included in the second user information, electronic device 100 can identify that the user of electronic device 100 is not suitable for using the neural network model included in the first external device 200-1.

[0134] As described above, user suitability is satisfied when the account information of the user of electronic device 100 matches the account information of the user of the first external device 200-1, but this disclosure is not limited thereto. That is, according to one embodiment, user suitability according to the invention can be satisfied when the user account of electronic device 100 has higher privileges than the user account of the first external device 200-1, and even when the account of the user of electronic device 100 is included in the same group as the account previously registered as the user of the first external device 200-1.

[0135] According to another embodiment, the external device receiving the first signal can also induce user authentication by sending a second signal (instead of sending second user information to the electronic device 100) that includes encrypted second device information and encrypted first model information. Specifically, if a password or biometric information (e.g., fingerprint information, iris information, etc.) is entered to decrypt the encrypted second device information and encrypted first model information, the electronic device 100 can identify that the user of the electronic device 100 is suitable to use the neural network model included in the first external device 200-1. There are no particular limitations on the encryption and decryption methods and the authentication process required for decryption.

[0136] refer to Figure 10 The user suitability identification process can be performed by the first external device 200-1 according to the present invention. Specifically, the user suitability identification process is not shown, but can be performed by the user suitability identification module 1300 included in the first external device 200-1.

[0137] refer to Figure 10 At operation S1010, electronic device 100 can receive first user input, and at operation S1020, it can send a first signal to first external device 200-1 based on the first user input to request information related to one or more neural network models included in one or more external devices.

[0138] In operation S1030, if a first signal is received, the first external device 200-1 can perform a user suitability identification process based on the first user information included in the first signal and the second user information stored in the first external device 200-1.

[0139] Specifically, if the account information of the user of electronic device 100 included in the first user information matches the account information of the user of the first external device 200-1 included in the second user information, the first external device 200-1 can identify that the user of electronic device 100 is suitable for using the neural network model included in the first external device 200-1. On the other hand, if the account information of the user of electronic device 100 included in the first user information does not match the account information of the user of the first external device 200-1 included in the second user information, the first external device 200-1 can identify that the user of electronic device 100 is not suitable for using the neural network model included in the first external device 200-1. If the first user information is not included in the first signal received from electronic device 100, the first external device 200-1 can receive the first user information by sending a request for the first user information to electronic device 100.

[0140] At operation S1040, if it is determined that the user of electronic device 100 is suitable for using the neural network model included in the first external device 200-1, the first external device 200-1 may send a second signal to electronic device 100, including second device information regarding the hardware specifications of the first external device 200-1 and first model information regarding one or more neural network models included in the first external device 200-1. Conversely, if it is determined that the user of electronic device 100 is not suitable for using the neural network model included in the first external device 200-1, the first external device 200-1 may not send the second signal to electronic device 100. Furthermore, if no second signal is received from the first external device 200-1 within a preset time, electronic device 100 may provide a user notification indicating that a request for information related to one or more neural network models included in the first external device 200-1 has been rejected.

[0141] As a result of performing the user suitability identification process as described above, if it is identified that the user of the electronic device 100 is suitable for using the neural network model included in the first external device 200-1, the electronic device 100 may perform the hardware suitability identification process and the model suitability identification process according to the present invention.

[0142] In other words, at operation S1050, if a second signal is received, the electronic device 100 can identify, based on the first device information, the second device information, and the first model information regarding the hardware specifications of the electronic device 100, at least one neural network model among the one or more neural network models included in the first external device 200-1 that is suitable for the hardware of the electronic device 100. Furthermore, at operation S1060, the electronic device 100 can, based on the first model information and the second model information regarding the one or more neural network models included in the electronic device 100, identify, among the one or more neural network models identified as suitable for the hardware of the electronic device 100, at least one neural network model suitable for replacing the neural network model included in the electronic device 100. The hardware suitability identification process and the model suitability identification process have been referenced. Figures 1 to 8 Detailed descriptions will be provided, therefore redundant descriptions of specific content will be omitted.

[0143] If the hardware suitability identification process and the model suitability identification process are performed, the electronic device 100 can receive second user input at operation S1070, and can request installation data of at least one neural network model selected based on the second user input from the first external device 200-1 at operation S1080. Furthermore, if installation data of the neural network model is requested, at operation S1090, the first external device 200-1 can send installation data of at least one selected neural network model to the electronic device 100.

[0144] The foregoing has described embodiments of the first external device 200-1 performing the user suitability identification process, the electronic device 100 performing the hardware suitability identification process, and the model suitability identification process, but the present invention is not limited thereto. That is, according to another embodiment of the present invention, the first external device 200-1 can also perform all the user suitability identification process, hardware suitability identification process, and model suitability identification process according to the present invention.

[0145] According to the above reference Figure 9 and 10 In the various embodiments described, it is possible to determine whether to transmit the neural network model only when it is identified that the user of the electronic device 100 is suitable to use the neural network model included in the external device through the user authentication process, and therefore, the security and reliability of the system can be improved during the transmission of the neural network model.

[0146] Figure 11 This is a diagram illustrating the user interface provided by an electronic device 100 according to an embodiment of the present invention.

[0147] refer to Figure 11According to the present invention, the electronic device 100 can display a user interface (UI) for receiving user input on its display. Furthermore, the electronic device 100 can receive user input for searching one or more neural network models included in one or more first external devices 200-1 via the user interface. For example, user input for searching at least one neural network model can be received based on user interaction in which a "search" UI element 2110 is selected from a plurality of UI elements included in the user interface.

[0148] Based on user input, electronic device 100 can send a first signal to request information related to one or more neural network models included in one or more first external devices 200-1. In response to the first signal, electronic device 100 can receive identification information about each of the one or more first external devices 200-1 and identification information included in each first external device 200-1. Furthermore, electronic device 100 can display the received identification information about the first external devices 200-1 and the identification information about the neural network models in a user interface. For example, electronic device 100 can display information indicating that first external device 200-1, referred to as device A, includes neural network models referred to as model A1, model A2, and model A3, and that first external device 200-1, referred to as device B, includes neural network models referred to as model B1 and model B2 in the user interface.

[0149] Specifically, the electronic device 100 can display UI elements in its user interface indicating whether each neural network model is a neural network model that satisfies the conversion suitability according to the invention, as well as identification information about the first external device 200-1 and identification information about the neural network model. For example, the electronic device 100 can display UI elements 2120 and 2130 in the form of "checkable checkboxes" to indicate that models A1 and A2 included in device A are neural network models that satisfy the conversion suitability. Furthermore, the electronic device 100 can display UI element 2140 in the form of "uncheckable checkboxes" to indicate that model A3 included in device A is a neural network model that does not satisfy the conversion suitability. In the case where models B1 and B2 are included in device B, the electronic device 100 can display UI element 2150 to indicate that the conversion suitability identification process according to the invention is "in progress".

[0150] After displaying a UI element indicating whether each neural network model is a neural network model that satisfies the transformation suitability according to the present invention, the electronic device 100 can receive user input for selecting one or more neural network models to be installed in the electronic device 100 from among one or more neural network models that satisfy the transformation suitability. For example, user input for selecting one or more neural network models to be installed in the electronic device 100 can be received based on user interaction of selecting a UI element 2130 from UI elements 2120 and 2130 displayed on the display. In this case, the electronic device 100 can, as Figure 11 UI element 2130 is shown as a "checkbox selected" to indicate that model A2 has been selected.

[0151] After selecting one or more neural network models to be installed in electronic device 100, if user input for installing the selected one or more neural network models is received, electronic device 100 can send a request for installation data of the one or more selected neural network models to each first external device 200-1 that includes the one or more selected neural network models. For example, user input for installing one or more selected neural network models in electronic device 100 can be received based on user interaction that selects the "Install" UI element 2160 among multiple UI elements included in the user interface.

[0152] In response to a request for installation data of one or more selected neural network models, electronic device 100 may receive installation data of one or more selected neural network models and install the corresponding neural network model in electronic device 100 based on the installation data.

[0153] The preceding text has described an embodiment in which an electronic device 100 displays a UI element indicating whether a neural network model satisfies transformation suitability. In this invention, satisfying transformation suitability means satisfying all the conditions required for each embodiment of hardware suitability, model suitability, and user suitability; and failing to satisfy transformation suitability means failing to satisfy at least some of the conditions required for each embodiment of hardware suitability, model suitability, and user suitability.

[0154] According to the above reference Figure 11 In the described embodiment, the electronic device 100 can improve user convenience and the efficiency of the process of transmitting the neural network model by simply requesting and receiving installation data of the neural network model selected by the user in the neural network model of the first external device 200-1 that satisfies the conversion suitability according to the present invention.

[0155] Figure 12 This is a diagram illustrating the user interface provided by the first external device 200-1 according to an embodiment of the present invention.

[0156] exist Figure 11 The user interface provided by the electronic device 100 has already been described, but according to another embodiment, the user interface for receiving user input according to the present invention can be displayed on the display of the first external device 200-1. Figure 12 The user interface shown represents a user interface for installing one or more applications included in the first external device 200-1 on the electronic device 100.

[0157] refer to Figure 12 The user interface may include UI elements representing applications included in the first external device 200-1. For example, such as... Figure 12 As shown, the user interface may include UI elements 2210 representing the "Call and Contacts" application, UI elements 2220 representing the "Messages" application, UI elements 2230 representing the "AI Secretary" application, UI elements 2250 representing the "Gallery" application, etc.

[0158] Furthermore, the user interface may include UI elements representing the neural network models included in each application within the first external device 200-1. (See again...) Figure 12 The user interface may include UI element 2240, which represents a neural network model called "AI Model X" included in the "AI Secretary" application, and UI element 2260, which represents a neural network model called "AI Model Y" included in the "gallery" application.

[0159] The first external device 200-1 can receive user input via a user interface for selecting one or more applications among a plurality of applications included in the first external device 200-1. For example, the first external device 200-1 can receive user input for selecting the "Call and Contacts" application, the "AI Secretary" application, and the "Gallery" application among the plurality of applications included in the first external device 200-1. Figure 12 In this context, UI elements in the form of "checkboxes" indicate that they are selected applications.

[0160] If user input for selecting one or more applications is received, the first external device 200-1 can send the installation data of one or more selected applications to the electronic device 100. For example, user input for selecting one or more applications can be received based on user interaction for selecting the "send" element 1270 from among multiple UI elements included in the user interface.

[0161] The first external device 200-1 can send second device information regarding its hardware specifications, first model information regarding the neural network model included in each of one or more selected applications, and installation data for the one or more selected applications to the user electronic device 100. Furthermore, the electronic device 100 can perform the hardware suitability identification process and model suitability identification process as described above based on the second device information received from the first external device 200-1, the first model information stored in the electronic device 100, and the first device information and second model information. Subsequently, if a request for installation data of a neural network model identified as satisfying hardware suitability and model suitability is received from the electronic device 100, the first external device 200-1 can send the identified neural network model's installation data to the electronic device 100.

[0162] At least some installation data of the application and installation data of the neural network model can be sent directly from the first external device 200-1 to the electronic device 100, and can also be sent to the electronic device 100 through a server that provides installation data of the application or installation data of the neural network model.

[0163] According to the above reference Figure 12 In the described embodiment, after performing the process of identifying the suitability of the neural network model included in the selected application, the installation data of the neural network model can be sent from the first external device 200-1 to the electronic device 100. While sending the installation data of the user-selected application from the first external device 200-1 to the electronic device 100, user convenience can be further improved. Specifically, Figure 12 The embodiments can be applied to the initial setup of an electronic device (new device) in which at least some information about the application and neural network model included in the first external device 200-1 (old device) is jointly sent to the electronic device 100 (new device).

[0164] Figure 13 This is a diagram illustrating an example of an electronic device, a first external device, and a second external device according to an embodiment of the present invention.

[0165] Figure 14 This is a sequence diagram illustrating the process by which an electronic device identifies a conversion suitability when a neural network model included in a first external device is transmitted to a second external device, according to an embodiment of the present invention.

[0166] refer to Figure 13According to the present invention, the electronic device 100 can be implemented as a smartphone, the first external device 200-1 can be implemented as a robotic cleaner, and the second external device 200-2 can be implemented as a smart companion robot. That is, the types of the electronic device 100, the first external device 200-1, and the second external device 200-2 according to the present invention can be different from each other. In this invention, the term "first external device 200-1" is used to specify an external device capable of sending installation data of a neural network model to the electronic device 100, while the term "second external device 200-2" is used to specify an external device capable of receiving installation data of a neural network model from the first external device 200-1.

[0167] On the other hand, when the neural network model included in the first external device 200-1 is transmitted to the second external device 200-2, because the first external device 200-1 and the second external device 200-2 do not include, Figure 13 When the monitor is shown, it may not display as described in the reference above. Figure 11 and 12 The user interface thus presents a problem in that it is difficult to receive user input for performing embodiments of the invention. Furthermore, when the first external device 200-1 and the second external device 200-2 do not include the software module according to the invention, there is a problem that the conversion suitability identification process according to the invention cannot be performed via either the first external device 200-1 or the second external device 200-2.

[0168] Therefore, according to embodiments of the present invention, the electronic device 100 can receive user input and can perform a hardware suitability identification process and a model suitability identification process based on information received from the first external device 200-1 and the second external device 200-2.

[0169] refer to Figure 14 In operation S1410, electronic device 100 can receive first user input. Furthermore, in operation S1420-1, electronic device 100 can, based on the first user input, send a request for first device information regarding the hardware specifications of first external device 200-1 and first model information regarding one or more neural network models included in first external device 200-1 to first external device 200-1. And in operation S1420-2, electronic device 100 can send a request for second device information regarding the hardware specifications of second external device 200-2 and second model information regarding one or more neural network models included in second external device 200-2 to second external device 200-2.

[0170] and Figures 1 to 12 The descriptions differ, in Figure 13 and 14In the description, the first device information is used as a term to specify information about the hardware specifications of the first external device 200-1, the second device information is used as a term to specify information about the hardware specifications of the second external device 200-2, the first model information is used as a term to specify information about one or more neural network models included in the first external device 200-1, and the second model information is used as a term to specify information about one or more neural network models included in the second external device 200-2.

[0171] In response to a request received from electronic device 100, at operation S1430-1, the first external device 200-1 can send the first device information and the first model information to electronic device 100, and at operation S1430-2, the second external device 200-2 can send the second device information and the second model information to electronic device 100.

[0172] If first device information and first model information are received from first external device 200-1, and second device information and second model information are received from second external device 200-2, at operation S1440, electronic device 100 can, based on the first device information, second device information, and first model information, identify one or more neural network models suitable for the hardware of second external device 200-2 from among one or more neural network models included in first external device 200-1. Furthermore, at operation S1450, electronic device 100 can, based on the first model information and second model information, identify one or more neural network models suitable for replacing the neural network models included in second external device 200-2 from among the one or more neural network models identified as suitable for the hardware of first external device 200-1. (Refer to...) Figure 13 and 14 In the described embodiments, the hardware suitability identification process and model suitability identification process described above can be applied similarly.

[0173] If the hardware suitability identification process and the model suitability identification process have been executed, at operation S1460, electronic device 100 can receive second user input. Furthermore, at operation S1470, electronic device 100 can send a request for installation data of one or more neural network models selected based on the second user input to first external device 200-1. If a request for installation data of one or more selected neural network models is received, at operation S1480, first external device 200-1 can send the installation data of one or more selected neural network models to second external device 200-2. First external device 200-1 can send installation data of one or more selected neural network models to electronic device 100, and electronic device 100 can also send the received installation data to second external device 200-2.

[0174] According to the above reference Figure 13 and 14 In the described embodiment, the electronic device 100 can be used as an intermediary device between the first external device 200-1 and the second external device, thereby improving the efficiency and reliability of the process of transmitting neural network models between external devices excluding displays or software modules according to the present invention.

[0175] Figure 15 This is a block diagram illustrating in detail the architecture of the software modules included in an electronic device 100 according to an embodiment of the present invention.

[0176] refer to Figure 15 The electronic device 100 according to the present invention may include software modules, such as a neural network model installation module 1500 and a neural network model performance monitor. Furthermore, the neural network model installation module may include a fitness identification module, a neural network model switching module 1510, and a neural network model fine-tuning module 1520, and the neural network model performance monitor 1700 may include a personalization level monitor 1710 and a hardware requirements specification monitor 1720.

[0177] "Fitness identification module" refers to a conceptual module that collectively refers to a module capable of identifying and converting suitability according to the present invention. Specifically, as described above, the suitability identification module may include a hardware suitability identification module 1100, a model suitability identification module 1200, and a user suitability identification module 1300. Furthermore, the hardware suitability identification module 1100, the model suitability identification module 1200, and the user suitability identification module 1300 may perform hardware suitability identification, model suitability identification, and user suitability identification processes based on first device information and second model information stored in the electronic device 100, and second device information and first model information received from an external device, according to different embodiments of the present invention. Figure 15The user information is shown in the form of being included in the first device information, but this is only relevant to this embodiment. (See above references.) Figures 1 to 14 The specific operations of the hardware suitability identification module 1100, the model suitability identification module 1200, and the user suitability identification module 1300 are described, therefore, redundant descriptions are omitted.

[0178] The "neural network model switching module" refers to a module that switches the neural network model included in the electronic device 100 to the neural network model included in the external device based on the installation data of the neural network model received from the external device. Specifically, the neural network model switching module can switch the neural network model included in the electronic device 100 to the neural network model included in the external device based on the structure and type of the neural network included in the neural network model, the number of layers in the neural network, the number of nodes in each layer, the weight value of each node, and the connection relationship between multiple nodes, as included in the installation data.

[0179] The “switching” from a neural network model included in electronic device 100 to a neural network model included in an external device can include replacing some neural network models of electronic device 100 with some neural network models of the external device, such as changing only the weight values ​​of nodes in the neural network model included in electronic device 100 to the weight values ​​of nodes in the neural network model included in the external device, and replacing the entire neural network model of electronic device 100 with the entire neural network model of the external device.

[0180] The "neural network model fine-tuning module" refers to a module that switches the neural network model included in the electronic device 100 to the neural network model included in the external device, and then fine-tunes the details. Specifically, the neural network model fine-tuning module can adjust the number of nodes in each layer and the weight value of each node in the neural network model by reflecting the detailed differences between the hardware specifications of the electronic device 100 and the hardware specifications of the external device, thereby making the neural network model more suitable for the electronic device 100. Furthermore, the neural network model fine-tuning module can adjust the parameters related to the personalization of the neural network model based on personalized data received from the external device. As mentioned above, "personalized data" refers to the training data used to personalize the neural network model included in the external device.

[0181] If the neural network model is switched / adjusted according to the neural network model switching module and the neural network model fine-tuning module, information about the neural network model can be stored / updated as second model information.

[0182] A "personalization level monitor" refers to a module that monitors the personalization level of the neural network model included in the electronic device 100. Specifically, when a second neural network model included in an external device is installed, and the first neural network model included in the electronic device 100 is personalized to some extent, the personalization level monitor can acquire information about the personalization level of the first neural network model and send the acquired information to the suitability identification module. Furthermore, the suitability identification module can identify whether the second neural network model is suitable to replace the first neural network model by comparing the information about the personalization level of the first neural network model received from the personalization level monitor with the information about the personalization level of the second neural network model received from the external device.

[0183] A "hardware requirements monitor" refers to a module used to monitor the hardware specifications required to execute the neural network models included in the electronic device 100. Specifically, the hardware requirements monitor can acquire information about the hardware specifications required to execute each of the one or more neural network models included in the electronic device 100, and this information can be stored as information included in the second model information. The "hardware specification information" included in the first device information is information about the hardware specifications of the electronic device 100 at the time of product release, or information about the hardware requirements specifications at the time of model suitability determination, which is different from the hardware requirements specification information of the neural network models acquired by the hardware requirements monitor.

[0184] The software modules included in the electronic device 100 have been described above, but this is only an embodiment of the invention. In addition to the modules shown, new configurations may be added or some modules may be omitted. Furthermore, at least two or more of the software modules according to the invention can be implemented as an integrated module.

[0185] Figure 16 This is a block diagram illustrating in detail the architecture of the software modules included in the first external device 200-1 according to an embodiment of the present invention.

[0186] refer to Figure 16 The first external device 200-1 may include software modules such as a neural network model installation module 1600 and a neural network model performance monitor 1700. Furthermore, the neural network model installation module 1600 may include a neural network model description information management module 1610 and a neural network model installation information management module 1620, and the neural network model performance monitor may include a personalization level monitor 1710 and a hardware requirements specification monitor 1720.

[0187] The "Neural Network Model Description Information Management Module 1610" refers to a module that comprehensively manages the information required to determine the conversion suitability of the neural network model included in the first external device 200-1. Specifically, the Neural Network Model Description Information Management Module 1610 can send the second device information and the first model information to the electronic device 100.

[0188] As described above, the second device information is information about the hardware specifications of the first external device 200-1, specifically referring to a term used broadly to refer to information about specifications indicating the performance characteristics of each piece of hardware included in the first external device 200-1. Furthermore, the first model information is information about one or more neural network models included in the first external device 200-1, and may specifically include information about service type, information about personalization level, and information about the hardware requirements specifications of each of the one or more neural network models included in the first external device 200-1.

[0189] In addition, the neural network model description information management module 1610 can collectively manage various information, such as the identification information, performance and version information of each neural network model included in the first external device 200-1.

[0190] The "Neural Network Model Installation Information Management Module 1620" refers to a module that manages the installation data of neural network models. Specifically, in response to an installation data request received from the electronic device 100, the Neural Network Model Installation Information Management Module 1620 can send the installation data of one or more neural network models to the electronic device 100. Specifically, the installation data may include the identification information of each of the one or more neural network models, and the configuration information required to install the one or more identified neural network models. The "configuration information" may include the structure and type of the neural network included in the neural network model, the number of layers included in the neural network, the number of nodes in each layer, the weight value of each node, and the connection relationships between multiple nodes.

[0191] In addition, the neural network model installation information management module 1620 can send personalized data of one or more neural network models along with installation data of one or more neural network models to the electronic device 100.

[0192] "Personalization Level Monitor 1710" refers to a module that monitors the personalization level of the neural network model included in the first external device 200-1. Specifically, the Personalization Level Monitor 1710 can output information about the personalization level of the neural network model included in the first external device 200-1 based on information about the user's history and information about the user's feedback. As mentioned above, the information about the user's history may include information about the number of times, frequency, and duration of the user's use of the neural network model, and the information about the user's feedback may include direct evaluation information on the usage results input by the user after using the neural network model and indirect evaluation information related to the usage results.

[0193] "Hardware Requirements Specification Monitor 1720" refers to a module that monitors the hardware specifications required to execute the neural network models included in the first external device 200-1. Specifically, the Hardware Requirements Specification Monitor 1720 can acquire information about the hardware specifications required to execute each of the one or more neural network models included in the first external device 200-1, and this information can be stored as information included in the first model information.

[0194] The software modules included in the first external device 200-1 have been described above, but this is only an embodiment of the invention. In addition to the modules shown, new configurations may be added or some modules may be omitted. Furthermore, at least two or more neural network models in the software modules according to the invention can be implemented as an integrated neural network model.

[0195] Figure 17 This is a block diagram schematically illustrating the architecture of a hardware configuration included in an electronic device according to an embodiment of the present invention.

[0196] Figure 18 This is a block diagram illustrating in more detail the architecture of the hardware configuration included in an electronic device according to an embodiment of the present invention.

[0197] refer to Figure 17 The electronic device 100 according to an embodiment of the present invention includes a communicator 110, a memory 120, and a processor 130. Furthermore, as... Figure 18 As shown, the electronic device 100 according to an embodiment of the present invention may further include a data acquisition unit 140 and a data output unit 150. However, as Figure 17 and 18 The configuration shown is merely an example; in implementing this invention, in addition to Figure 17 and Figure 18 In addition to the configuration shown, you can add new configurations or omit some configurations.

[0198] The communicator 110 may include circuitry and communicate with external devices. Specifically, the processor 130 may receive various data or information from external devices connected via the communicator 110, and may also send various data or information to external devices.

[0199] The communicator 110 may include at least one of a Wi-Fi module, a Bluetooth module, a wireless communication module, or a near-field communication (NFC) module. Specifically, each of the Wi-Fi and Bluetooth modules can perform communication in both Wi-Fi and Bluetooth modes. When using a Wi-Fi or Bluetooth module, various connection information, such as a Service Set Identifier (SSID), is first sent and received to establish communication, and then various information can be sent and received.

[0200] Furthermore, the wireless communication module can perform communication according to various communication protocols, such as the Institute of Electrical and Electronics Engineers (IEEE), Zigbee, 3G, 3GPP, LTE, and 5G. Additionally, the NFC module can use the 13.56MHz band from various radio frequency identification (RF-ID) bands (such as 135kHz, 13.56MHz, 433MHz, 860-960MHz, and 2.45GHz) to perform communication via NFC.

[0201] Specifically, in different embodiments of the present invention, the communicator 110 may send to the first external device 200-1 a first signal requesting information related to one or more neural network models included in the first external device 200-1, and a third signal requesting installation data for one or more neural network models. Furthermore, in response to the first signal, the communicator 110 may receive from the first external device 200-1 a second signal including second device information comprising the hardware specifications of the first external device 200-1 and first model information comprising one or more neural network models included in the first external device 200-1, and in response to the third signal, receive a fourth signal including installation data for one or more identified neural network models.

[0202] One or more instructions concerning the electronic device 100 may be stored in the memory 120. Furthermore, the memory 120 may store an operating system (O / S) for driving the electronic device 100. Additionally, the memory 120 may also store various software programs or application programs for operating the electronic device 100 according to various embodiments of the present invention. Furthermore, the memory 120 may include a semiconductor memory such as flash memory, or a magnetic storage medium such as a hard disk.

[0203] Specifically, the memory 120 can store various software modules for operating the electronic device 100 according to different embodiments of the present invention, and the processor 130 can execute the various software modules stored in the memory 120 to control the operation of the electronic device 100. That is, the memory 120 is accessed by the processor 130, and the reading, writing, correction, deletion, updating, etc. of data in the memory 120 can be performed by the processor 130.

[0204] On the other hand, in this invention, the term memory 120 can be used to mean including memory 120, read-only memory (ROM) (not shown), random access memory (RAM) (not shown) in processor 130, or a memory card (not shown) (e.g., a micro-secure digital (SD) card or memory stick) installed in electronic device 100.

[0205] Specifically, in various embodiments of the present invention, memory 120 may store first device information regarding the hardware specifications of electronic device 100, and second model information regarding one or more neural network models included in electronic device 100. Furthermore, memory 120 may store information as referenced above. Figure 15 The various modules include a hardware fitness identification module 1100, a model fitness identification module 1200, and a user fitness identification module 1300. Furthermore, various information required to achieve the objectives of this invention can be stored in the memory 120, and the information stored in the memory 120 can be updated when received from an external device or by user input.

[0206] The processor 130 controls the overall operation of the electronic device 100. Specifically, the processor 130 can be connected to the configuration of the electronic device 100, including the communicator 110 and the memory 120, as described above, and can execute one or more instructions stored in the memory 120 to control the overall operation of the electronic device 100 as described above.

[0207] Processor 130 can be implemented in various ways. For example, processor 130 can be implemented as at least one of application-specific integrated circuit (ASIC), embedded processor, microprocessor, hardware control logic, hardware finite state machine (FSM), or digital signal processor (DSP). On the other hand, in this invention, the term processor 130 can be used to mean including central processing unit (CPU), graphics processing unit (GPU), main processing unit (MPU), etc.

[0208] Specifically, in different embodiments of the present invention, when user input is received, the processor 130 can control the communicator 110 to send a first signal to request information related to one or more neural network models included in one or more external devices. The processor 130 can receive a second signal from a first external device 200-1 among the one or more external devices as a response to the first signal. This second signal includes second device information regarding the hardware specifications of the first external device 200-1 and first model information regarding one or more neural network models included in the first external device 200-1. The processor 130 can receive a second signal from a first external device 200-1 among the one or more external devices as a response to the first signal. This second signal includes second device information regarding the hardware specifications of the first external device 200-1 and first model information regarding the one or more neural network models included in the first external device 200-1. Model information is input to the hardware suitability identification module 1100 to identify whether each of the one or more neural network models included in the first external device 200-1 is suitable for the hardware of the electronic device 100. The communicator 110 can be controlled to send a third signal to the first external device 200-1, the third signal including a request for installation data of the one or more neural network models identified as suitable for the hardware of the electronic device 100. The communicator 110 can also receive a fourth signal from the first external device 200-1 as a response to the third signal, the fourth signal including installation data of the one or more neural network models identified as suitable for the hardware of the electronic device 100.

[0209] Furthermore, if one or more neural network models suitable for the hardware of electronic device 100 are identified, processor 130 can also identify whether each of the one or more neural network models identified as suitable for the hardware of electronic device 100 is suitable to replace the neural network model included in electronic device 100 by inputting first model information and second model information into model suitability identification module 1200. (See above reference) Figures 1 to 16 Various embodiments of the present invention, based on the control of the processor 130, have been described, and therefore redundant descriptions will be omitted.

[0210] The data acquisition device 140 may include circuitry, through which the processor 130 can acquire various types of data used in the electronic device 100. Specifically, the data acquisition device 140 may include a camera 141, a microphone 142, a sensor 143, etc.

[0211] Camera 141 can acquire an image of at least one object. Specifically, camera 141 may include an image sensor, which can convert light entering through a lens into an electrical image signal. Furthermore, microphone 142 can receive voice signals and convert the received voice signals into electrical signals.

[0212] Sensor 143 can detect various information both inside and outside the electronic device 100. Specifically, sensor 143 may include at least one of a Global Positioning System (GPS) sensor, a gyroscope sensor, an accelerometer, a lidar sensor, an inertial sensor (inertial measurement unit (IMU)), or a motion sensor. Furthermore, sensor 143 may include various types of sensors, such as temperature sensors, humidity sensors, infrared sensors, and biosensors.

[0213] Specifically, in various embodiments of the invention, the data acquisition unit 140 can acquire data input to one or more neural network models included in the electronic device 100. For example, the processor 130 can acquire image data input to a neural network model related to object recognition via a camera 141, and can acquire speech signals input to a neural network model related to speech recognition via a microphone 142. Furthermore, the processor 130 can also acquire location information input to a neural network model related to autonomous driving via at least one sensor 143 of a GPS sensor or a lidar sensor.

[0214] The data output device 150 may include circuitry, and the processor 130 may output various functions that the electronic device 100 can perform through the data output device 150. Specifically, the data output device 150 may include at least one of a display 151, a speaker 152, or an indicator 153.

[0215] Display 151 can output image data. Specifically, display 151 can display images or user interfaces stored in memory 120 under the control of processor 130. Display 151 can be implemented as a liquid crystal display (LCD) panel, organic light-emitting diode (OLED), etc., and in some cases, it can also be implemented as a flexible display, transparent display, etc. However, the display 151 according to the present invention is not limited to a specific type. Display 151 can be implemented as a touch display and configured to receive touch interaction from the user. Speaker 152 can output audio data under the control of processor 130, and indicator 153 can be lit under the control of processor 130.

[0216] Specifically, in various embodiments of the invention, the display 151 may display a user interface including information about neural network models that satisfy the hardware suitability and model suitability requirements according to the invention. Furthermore, the processor 130 may receive user input via the user interface for selecting one or more neural network models among one or more neural network models.

[0217] If it is identified that none of the neural network models included in the external device satisfy both hardware suitability and model suitability, the processor 130 may also output a user notification to indicate that none of the neural network models included in the external device satisfy the transformation suitability according to the invention. The user notification may be output as visual information via the display 151, and as audio information via the speaker 152, or as an indicator 153 being illuminated.

[0218] On the other hand, the control method of the electronic device 100 according to the above embodiments can be implemented by a program and provided to the electronic device 100. Specifically, the program including the control method of the electronic device 100 can be stored in a non-transitory computer-readable medium and provided.

[0219] Specifically, in a non-transitory computer-readable recording medium including a program for executing a control method of electronic device 100, the control method of electronic device 100 may include, upon receiving user input, sending a first signal for requesting information related to one or more neural network models included in one or more external devices, and receiving a second signal via communicator 110 from a first external device 200-1 among the one or more external devices as a response to the first signal. The second signal includes second device information regarding the hardware specifications of the first external device 200-1 and first model information regarding the one or more neural network models included in the first external device 200-1. By inputting the first device information, the second device information, and the first model information into the hardware suitability identification module 1100, the module identifies whether each of the one or more neural network models included in the first external device 200-1 is suitable for the hardware of the electronic device 100. A third signal including a request for installation data of the one or more neural network models identified as suitable for the hardware of the electronic device 100 is sent to the first external device 200-1. In response to the third signal, a fourth signal including the installation data of the one or more neural network models identified as suitable for the hardware of the electronic device 100 is received from the first external device 200-1 through the communicator 110.

[0220] Furthermore, identifying whether each of the one or more neural network models included in the first external device 200-1 is suitable for the hardware of the electronic device 100 may include identifying the one or more neural network models included in the first external device 200-1 as suitable for the hardware of the electronic device 100 when the specifications of each of the plurality of hardware configurations included in the electronic device 100 are greater than or equal to the specifications of the plurality of hardware configurations included in the first external device 200-1, and identifying one or more neural network models included in the first external device 200-1 whose hardware requirement specifications are lower than the specifications of the plurality of hardware configurations included in the electronic device 100 as suitable for the hardware of the electronic device when the specifications of one or more of the plurality of hardware configurations included in the electronic device 100 are lower than the specifications of the plurality of hardware configurations included in the first external device 200-1.

[0221] Non-transitory computer-readable media are not short-term storage media such as registers, caches, or memory chips, but rather machine-readable media that store data semi-permanently. Specifically, the various applications or programs mentioned above can be stored and provided on non-transitory computer-readable media, such as optical discs (CDs), digital multifunction discs (DVDs), hard disks, Blu-ray discs, Universal Serial Bus (USB), memory cards, and read-only memory (ROM).

[0222] The control method of electronic device 100 and the computer-readable recording medium including the program for executing the control method of electronic device 100 have been briefly described above, but this is only to omit redundant descriptions, and various embodiments of electronic device 100 can also be applied to the control method of electronic device 100 and the computer-readable recording medium including the program for executing the control method of electronic device 100.

[0223] According to the various embodiments described above, when transmitting a neural network model from an external device, the electronic device 100 can determine whether to transmit the neural network model by identifying whether it is suitable to transmit the neural network model to the electronic device 100, and accordingly, the efficiency and reliability of transmitting the learning may be significantly improved.

[0224] Functions related to the aforementioned neural network model can be executed via memory 120 and processor 130. Processor 130 can be configured as one or more processors. In this case, the one or more processors 130 can be a general-purpose processor 130 such as a CPU, AP, etc., a graphics-specific processor 130 such as a GPU, VPU, etc., or an artificial intelligence-specific processor such as an NPU. The one or more processors 130 perform control to process input data according to predefined operating rules or artificial intelligence models stored in non-volatile memory 120 and volatile memory 120. The predefined operating rules or artificial intelligence models are characterized by being created through training.

[0225] Here, "created through training" refers to a predefined set of operating rules or an artificial intelligence model with expected characteristics created by applying a learning algorithm to a large amount of learning data. This learning can be performed within the device itself that performs the artificial intelligence according to the invention, or it can be performed via a separate server / system.

[0226] Artificial intelligence models may include multiple neural network layers. Each layer has multiple weight values, and layer computation is performed by calculating the results of the previous layer and the multiple weight values. Examples of neural networks include convolutional neural networks (CNNs), deep neural networks (DNNs), recurrent neural networks (RNNs), restricted Boltzmann machines (RBMs), deep belief networks (DBNs), bidirectional recurrent deep neural networks (BRDNNs), generative adversarial networks (GANs), and deep Q-networks. Unless otherwise specified, the neural networks in this invention are not limited to the examples mentioned above.

[0227] A learning algorithm is a method that uses a large amount of learning data to train a predetermined target device (such as a robot) so that the target device can make decisions or predict itself. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning. The learning algorithms in this invention are not limited to the above examples, except where specified.

[0228] Machine-readable storage media may be provided in the form of non-transitory storage media. The term "non-transitory" means only that the storage medium is a tangible device and does not include signals (e.g., electromagnetic waves), and the term does not distinguish between cases where data is stored semi-permanently in the storage medium and cases where data is temporarily stored in the storage medium. For example, "non-transitory storage media" may include buffers for temporarily storing data.

[0229] According to embodiments, methods based on different embodiments of the invention may be included and provided in a computer program product. The computer program product can be traded as a product between a seller and a buyer. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., an optical disc read-only memory (CD-ROM)), or distributed online (e.g., downloaded or uploaded) through an app store (e.g., the Play Store™), or distributed directly between two user devices (e.g., smartphones) (e.g., downloaded or uploaded). In the case of online distribution, at least a portion of the computer program product (e.g., a downloadable application) may be stored at least temporarily in a machine-readable storage medium, such as the memory of a manufacturer's server, an app store's server, or a relay server, or may be temporarily generated.

[0230] Each component (e.g., module or program) according to the different embodiments described above may include a single entity or multiple entities, and some sub-components of the aforementioned sub-components may be omitted, or other sub-components may be included in different embodiments. Optionally or additionally, some components (e.g., module or program) may be integrated into one entity to perform the same or similar functions performed by the individual components prior to integration.

[0231] According to different embodiments, operations performed by modules, programs or other components may be performed sequentially, in parallel, iteratively or heuristically, or at least some operations may be performed in a different order or omitted, or other operations may be added.

[0232] On the other hand, the terms "device" or "module" used in this invention include units composed of hardware, software, or firmware, and are used interchangeably with terms such as logic, logic block, component, or circuit. A "device" or "module" can be a component formed as a whole or the smallest unit performing one or more functions or a portion thereof. For example, the module can be configured as an application-specific integrated circuit (ASIC).

[0233] Various embodiments of the present invention can be implemented in software, including instructions stored in a machine-readable storage medium (e.g., a computer). The machine is a device that invokes the stored instructions from the storage medium and is operable according to the invoked instructions, and may include electronic devices (e.g., electronic device 100) according to the disclosed embodiments.

[0234] When an instruction is executed by processor 130, processor 130 may directly or using other components under the control of processor 130 execute the function corresponding to the instruction. Instructions may include code generated or executed by a compiler or interpreter.

[0235] Although the invention has been shown and described with reference to various embodiments thereof, those skilled in the art will understand that various changes to the form and details of the invention may be made without departing from the spirit and scope of the invention as defined by the appended claims and their equivalents.

Claims

1. An electronic device for identifying the transformation suitability of a neural network model included in an external device, the electronic device comprising: communicator; The memory is configured to store first device information about the hardware specifications of the electronic device, and a hardware suitability recognizer that identifies a neural network model suitable for the hardware of the electronic device. as well as The processor is configured as follows: Based on the received user input, the communicator is controlled to send a first signal to request information related to one or more neural network models included in one or more external devices. In response to the first signal, a second signal is received via the communicator from a first external device among the one or more external devices. The second signal includes second device information regarding the hardware specifications of the first external device and first model information regarding one or more neural network models included in the first external device. By inputting the first device information, the second device information, and the first model information into a hardware suitability identifier, it is determined whether each of the one or more neural network models included in the first external device is suitable for the hardware of the electronic device. The communicator is controlled to send a third signal to the first external device, the third signal including a request for installation data of one or more neural network models identified as suitable for the hardware of the electronic device, and In response to the third signal, a fourth signal is received from the first external device via the communicator, the fourth signal including installation data of the one or more neural network models identified as suitable for the hardware of the electronic device, wherein the processor is further configured to: Based on the fact that the specifications of each of the plurality of hardware configurations included in the electronic device are greater than or equal to the specifications of the plurality of hardware configurations included in the first external device, the one or more neural network models included in the first external device are identified as suitable for the hardware of the electronic device, and Based on the fact that the specifications of one or more of the plurality of hardware configurations included in the electronic device are smaller than the specifications of the plurality of hardware configurations included in the first external device, one or more neural network models included in the first external device that have hardware requirement specifications lower than the specifications of the plurality of hardware configurations included in the electronic device are identified as hardware suitable for the electronic device.

2. The electronic device according to claim 1, in, The first device information includes specifications of the processor, the memory, and the data acquisition device included in the electronic device. The device includes a data acquisition unit that acquires data input to one or more neural network models included in the electronic device, and includes at least one of a camera, microphone, or sensor included in the electronic device. The second device information includes specifications of the processor included in the first external device, specifications of the memory included in the first external device, and specifications of the data acquisition device included in the first external device. The data acquisition device included in the first external device acquires data input to the one or more neural network models included in the first external device, and includes at least one of a camera, microphone, or sensor included in the first external device. The first model information includes information on the hardware requirements specifications for each of the one or more neural network models included in the first external device.

3. The electronic device according to claim 2, in, The memory stores second model information about the one or more neural network models included in the electronic device, and a model fitness identifier. The processor is further configured to, based on the identification of one or more neural network models suitable for the hardware of the electronic device, identify whether each of the one or more neural network models identified as suitable for the hardware of the electronic device is suitable for replacing the neural network model included in the electronic device by inputting the first model information and the second model information into a model suitability recognizer. The third signal includes a request for installation data of one or more neural network models identified as suitable for replacing a neural network model included in the electronic device, and The fourth signal includes installation data that is identified as suitable for replacing one or more neural network models included in the electronic device.

4. The electronic device according to claim 3, in, The first model information also includes information about the service type and information about the level of personalization for each of the one or more neural network models included in the first external device, and The second model information also includes information about the service type and information about the level of personalization of each of the one or more neural network models included in the electronic device.

5. The electronic device according to claim 4, wherein, The processor is also configured to: Based on the service type information included in the first model information and the second model information, the service types of the one or more neural network models included in the electronic device are compared with the service types of the one or more neural network models included in the first external device identified as suitable for the hardware of the electronic device. Based on the fact that the service type of a first neural network model included in one or more neural network models in the first external device identified as suitable for the hardware of the electronic device is the same as the service type of a second neural network model included in one or more neural network models in the electronic device, and based on information about the level of personalization included in each of the first model information and the second model information, the personalization level of the first neural network model and the personalization level of the second neural network model are compared, and... Based on the fact that the personalization level of the first neural network model is higher than that of the second neural network model, it is determined that the first neural network model is suitable to replace the second neural network model.

6. The electronic device according to claim 5, in, The personalization level of the first neural network model is identified based on at least one of information regarding the user's usage history of the first external device and information regarding user feedback on the first external device. The personalization level of the second neural network model is identified based on at least one of information about the user's usage history of the electronic device and information about the user's feedback on the electronic device.

7. The electronic device according to claim 1, in, When user authentication is completed based on first user information about the electronic device and second user information about the first external device, the second signal is received from the first external device via the communicator. Wherein, the first user information includes at least one of user account information or identification information of the electronic device, and The second user information includes at least one of the user's account information or the identification information of the first external device.

8. The electronic device according to claim 1, further comprising a display, wherein, The processor is also configured to: The display is controlled to show a user interface, the user interface including information about the one or more neural network models identified as suitable for the hardware of the electronic device, and User input is received through the user interface for selecting one or more neural network models from among the one or more neural network models identified as suitable for the hardware of the electronic device. The third signal includes a request for installation data for the one or more selected neural network models, and The fourth signal includes installation data for the one or more selected neural network models.

9. A control method for an electronic device, the electronic device storing first device information about the hardware specifications of the electronic device and a hardware suitability recognizer for identifying a neural network model suitable for the hardware of the electronic device, and recognizing the transformation suitability of a neural network model included in an external device, the control method comprising: Based on the received user input, a first signal is sent to request information related to one or more neural network models included in one or more external devices; In response to the first signal, a second signal is received from a first external device among the one or more external devices, the second signal including second device information about the hardware specifications of the first external device and first model information about one or more neural network models included in the first external device; By inputting the first device information, the second device information, and the first model information into the hardware suitability identifier, it is identified whether each of the one or more neural network models included in the first external device is suitable for the hardware of the electronic device; Sending a third signal to the first external device, the third signal including a request for installation data for one or more neural network models identified as suitable for the hardware of the electronic device; and In response to the third signal, a fourth signal is received from the first external device, the fourth signal including installation data of the one or more neural network models identified as suitable for the hardware of the electronic device, wherein identifying whether each of the one or more neural network models is suitable for the hardware of the electronic device includes: Based on the fact that the specifications of each of the plurality of hardware configurations included in the electronic device are greater than or equal to the specifications of the plurality of hardware configurations included in the first external device, the one or more neural network models included in the first external device are identified as hardware suitable for the electronic device. Based on the fact that the specifications of one or more of the plurality of hardware configurations included in the electronic device are smaller than the specifications of the plurality of hardware configurations included in the first external device, one or more neural network models included in the first external device that have hardware requirement specifications lower than the specifications of the plurality of hardware configurations included in the electronic device are identified as hardware suitable for the electronic device.

10. The control method according to claim 9, in, The first device information includes specifications regarding the processor included in the electronic device, the memory included in the electronic device, and the data acquisition device included in the electronic device. The device includes a data acquisition unit that acquires data input to one or more neural network models included in the electronic device, and includes at least one of a camera, microphone, or sensor included in the electronic device. The second device information includes specifications of the processor included in the first external device, specifications of the memory included in the first external device, and specifications of the data acquisition device included in the first external device. The data acquisition device included in the first external device acquires data input to the one or more neural network models included in the first external device, and includes at least one of a camera, microphone, or sensor included in the first external device. The first model information includes information on the hardware requirements specifications for each of the one or more neural network models included in the first external device.

11. The control method according to claim 10, in, The electronic device also stores second model information about the one or more neural network models included in the electronic device, and a model fitness recognizer. The control method further includes identifying, based on the identification of one or more neural network models suitable for the hardware of the electronic device, whether each of the one or more neural network models identified as suitable for the hardware of the electronic device is suitable for replacing the neural network model included in the electronic device by inputting the first model information and the second model information into the model suitability recognizer. The third signal includes a request for installation data of one or more neural network models identified as suitable for replacing a neural network model included in the electronic device, and The fourth signal includes installation data that is identified as suitable for replacing one or more neural network models included in the electronic device.

12. The control method according to claim 11, in, The first model information also includes information about the service type and information about the level of personalization for each of the one or more neural network models included in the first external device, and The second model information also includes information about the service type and information about the level of personalization of each of the one or more neural network models included in the electronic device.

13. The control method according to claim 12, wherein, Identifying one or more neural network models suitable for replacing neural network models included in the electronic device includes: Based on the service type information included in the first model information and the second model information, the service types of the one or more neural network models included in the electronic device are compared with the service types of the one or more neural network models included in the first external device identified as suitable for the hardware of the electronic device. The service type of a first neural network model in one or more neural network models identified as suitable for the hardware of the electronic device is the same as the service type of a second neural network model included in one or more neural network models in the electronic device. Based on information about the level of personalization included in each of the first model information and the second model information, the personalization level of the first neural network model and the personalization level of the second neural network model are compared. Based on the fact that the personalization level of the first neural network model is higher than that of the second neural network model, it is determined that the first neural network model is suitable to replace the second neural network model.

14. A non-transitory computer-readable recording medium comprising a program for performing a control method for an electronic device, the electronic device storing first device information about the hardware specifications of the electronic device, a hardware fitness identifier for identifying a neural network model suitable for the hardware of the electronic device, and a transformation fitness identifier for identifying a neural network model included in an external device, wherein the control method comprises: Based on the received user input, a first signal is sent to request information related to one or more neural network models included in one or more external devices; In response to the first signal, a second signal is received from a first external device among the one or more external devices, the second signal including second device information about the hardware specifications of the first external device and first model information about one or more neural network models included in the first external device; By inputting the first device information, the second device information, and the first model information into the hardware suitability identifier, it is identified whether each of the one or more neural network models included in the first external device is suitable for the hardware of the electronic device; Sending a third signal to the first external device, the third signal including a request for installation data for one or more neural network models identified as suitable for the hardware of the electronic device; and In response to the third signal, a fourth signal is received from the first external device, the fourth signal including installation data of the one or more neural network models identified as suitable for the hardware of the electronic device. The hardware for identifying whether each of the one or more neural network models is suitable for the electronic device includes: Based on the fact that the specifications of each of the plurality of hardware configurations included in the electronic device are greater than or equal to the specifications of the plurality of hardware configurations included in the first external device, the one or more neural network models included in the first external device are identified as hardware suitable for the electronic device. Based on the fact that the specifications of one or more of the plurality of hardware configurations included in the electronic device are smaller than the specifications of the plurality of hardware configurations included in the first external device, one or more neural network models included in the first external device that have hardware requirement specifications lower than the specifications of the plurality of hardware configurations included in the electronic device are identified as hardware suitable for the electronic device.

15. An electronic device for identifying the transformation suitability of a neural network model included in an external device, the electronic device comprising: communicator; The memory is configured to store internal model information about one or more neural network models included in the electronic device, as well as a model fitness recognizer; as well as The processor is configured as follows: Based on the received user input, the communicator is controlled to send a first signal to request information related to one or more neural network models included in one or more external devices. In response to the first signal, a second signal is received via the communicator from a first external device among the one or more external devices, regarding external model information about one or more neural network models included in the first external device. By inputting the internal model information and the external model information into a model suitability identifier, it is determined whether each of the one or more neural network models included in the first external device is suitable for replacing the one or more neural network models included in the electronic device. The communicator is controlled to send a third signal to the first external device, the third signal including a request for installation data of one or more neural network models identified as suitable for replacing a neural network model included in the electronic device, and In response to the third signal, a fourth signal is received from the first external device via the communicator, the fourth signal including the installation data of the one or more identified neural network models, wherein the processor is further configured to: Based on the service type information included in each of the internal model information and the external model information, the service types of the one or more neural network models included in the electronic device are compared with the service types of the one or more neural network models included in the first external device. Based on the fact that the service type of the first neural network model included in the one or more neural network models in the first external device is the same as the service type of the second neural network model included in the one or more neural network models in the electronic device, and based on information about the personalization level included in each of the internal model information and the external model information, the personalization level of the first neural network model and the personalization level of the second neural network model are compared, and Based on the fact that the personalization level of the first neural network model is higher than that of the second neural network model, the first neural network model is identified as suitable for replacing the second neural network model.

Images