Classifications

• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F11/00—Error detection; Error correction; Monitoring
  • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
  • G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
  • G06F11/079—Root cause analysis, i.e. error or fault diagnosis
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F18/00—Pattern recognition
  • G06F18/30—Post-processing
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
  • G06N20/00—Machine learning

Definitions

• This disclosure relates to an information processing device, an information processing method, and a program.
• Predictive models generated by machine learning are used to predict the output for new input data.
• prediction errors can occur due to various factors. In such cases, analyzing the factors behind the prediction errors is important in order to improve the predictive model.
• Patent Document 1 discloses a technology for analyzing the causes of prediction errors in a prediction model.
• indexes are calculated for explanatory variables or objective variables used in the prediction model, and the causes of the prediction errors are identified using this.
• Patent Document 1 analyzes the causes of prediction errors by evaluating the degree of anomaly in explanatory variables and evaluating the distribution distance between training data and operational data.
• Patent Document 1 only analyzes whether the cause of a prediction error by a prediction model is due to the sample, and is unable to quantitatively evaluate the cause. This makes it difficult to consider appropriate measures to improve the prediction model depending on the cause of the prediction error, resulting in the problem that it is not possible to further improve the accuracy of the prediction model.
• the objective of this disclosure is to provide an information processing device that can solve the above-mentioned problem of being unable to further improve the accuracy of the predictive model.
• An information processing device includes: an error calculation unit that calculates a prediction error between a predicted value, which is an output when an explanatory variable of the target data is input to the prediction model, and a response variable of the target data; an index calculation unit that calculates an index for evaluating a contribution to the prediction error in at least one of the explanatory variables of the target data, the objective variable of the target data, and the prediction model, based on data that can be used to calculate the prediction error; a contribution calculation unit that calculates the contribution amount based on the prediction error and the index; Equipped with The structure is as follows.
• an information processing method includes: Calculating a prediction error between a predicted value, which is an output when an explanatory variable of the target data is input to a prediction model, and a response variable of the target data; Calculating an index for evaluating a contribution to the prediction error in at least one of the explanatory variables of the target data, the objective variable of the target data, and the prediction model, based on data that can be used to calculate the prediction error; Calculating the contribution amount based on the prediction error and the index.
• the structure is as follows.
• a program includes: Calculating a prediction error between a predicted value, which is an output when an explanatory variable of the target data is input to a prediction model, and a response variable of the target data; Calculating an index for evaluating a contribution to the prediction error in at least one of the explanatory variables of the target data, the objective variable of the target data, and the prediction model, based on data that can be used to calculate the prediction error; Calculating the contribution amount based on the prediction error and the index.
• the structure is as follows.
• FIG. 2 is a diagram illustrating functions of an information processing device according to the first embodiment of the present disclosure.
• 1 is a block diagram showing a configuration of an information processing device according to a first embodiment of the present disclosure.
• FIG. 3 is a diagram showing the contents of processing by the information processing device disclosed in FIG. 2 .
• 3 is a flowchart showing the operation of the information processing device disclosed in FIG. 2 .
• FIG. 11 is a block diagram showing a configuration of an information processing device according to a third embodiment of the present disclosure.
• FIG. 6 is a diagram showing the contents of processing by the information processing device disclosed in FIG. 5 .
• FIG. 11 is a block diagram showing a hardware configuration of an information processing device according to a fourth embodiment of the present disclosure.
• FIG. 11 is a block diagram showing a configuration of an information processing device according to a fourth embodiment of the present disclosure.
• Fig. 1 is a diagram for explaining the function of an information processing device.
• Fig. 2 is a diagram for explaining the configuration of the information processing device, and
• Figs. 3 and 4 are diagrams for explaining the processing operation of the information processing device.
• the information processing device 10 in this embodiment quantitatively evaluates the factors of prediction error in a prediction model generated by machine learning. For example, as shown by the arrow in FIG. 1, it is assumed that there is a prediction error in the predicted value, which is the output when sample data is input to the prediction model. In this case, the information processing device 10 calculates the factors of the prediction error by breaking them down into the contribution amounts of each data. Note that in this embodiment, as shown in FIG. 1, an example will be described in which the contribution amounts of the factors of the prediction error are quantitatively calculated by breaking them down into the contribution amount of the explanatory variable of the sample data, the contribution amount of the objective variable of the sample data, and the contribution amount of the prediction model. However, the information processing device 10 is not limited to calculating all of the above-mentioned contribution amounts, and may calculate at least one of them.
• the information processing device 10 is composed of one or more information processing devices each having a calculation device and a storage device. As shown in FIG. 2, the information processing device 10 is equipped with a decomposition target sample input unit 11, a decomposition result output unit 12, an error calculation unit 13, an index calculation unit 14, and an error decomposition unit 15. The functions of the decomposition target sample input unit 11, the decomposition result output unit 12, the error calculation unit 13, the index calculation unit 14, and the error decomposition unit 15 can be realized by the calculation device executing a program for realizing each function stored in the storage device.
• the information processing device 10 is also equipped with a prediction model storage unit 16 and a reference data storage unit 17.
• the prediction model storage unit 16 and the reference data storage unit 17 are composed of a storage device. Each component will be described in detail below.
• the prediction model storage unit 16 stores data constituting a prediction model f that has been generated in advance by machine learning.
• the prediction model f is, for example, a machine learning model generated by performing supervised learning, and is generated by supervised learning of training data (X, Y) consisting of a set of explanatory variable X and target variable Y, and is configured to output a predicted value by inputting unknown explanatory variables.
• the reference data storage unit 17 stores reference data D that can be input to the prediction model f.
• the reference data D is data consisting of a set (X, Y) of explanatory variables X and objective variables Y, and is data that can be used in the prediction model f, such as training data used when learning the prediction model f, verification data or evaluation data used when evaluating the prediction model f, operation data used when operating the prediction model f, or the same type of data as the training data or operation data that can be input to the prediction model f.
• the reference data D is training data.
• the decomposition target sample input unit 11 accepts input of sample data (target data) to be decomposed into factors of prediction errors.
• the sample data is data consisting of a set (x * , y * ) of explanatory variable x * and objective variable y * , and may be data included in the above-mentioned reference data D or may be data not included in the reference data D.
• the decomposition result output unit 12 outputs the contribution of each data to the prediction error. For example, the decomposition result output unit 12 outputs the contribution of the explanatory variables of the sample data, the contribution of the objective variables of the sample data, and the contribution of the prediction model calculated by the error decomposition unit 15, as described below, by displaying them on a display device.
• the prediction error can be expressed as the squared error (y * -f(x * )) 2 between the predicted value f(x * ) and the objective variable y * of the sample data.
• the prediction error may be expressed by any loss function such as a residual error, a 0-1 loss, or the like .
• the index calculation unit 14 calculates an index s * for evaluating the contribution of each data to the prediction error.
• an index s * x for the explanatory variable of the sample data an index s * y for the objective variable of the sample data, and an index s * f for the prediction model are calculated.
• each index s * is calculated based on data that can be used to calculate the prediction error, including data that can be used in the prediction model f, such as the prediction model f, the reference data D, and the sample data (x * , y*).
• each index s * can be calculated based on data used in the generation, evaluation, and operation of the prediction model f and data that can be input to the prediction model f.
• an example of the index s * will be described with reference to FIG. 3. Note that the index s * is calculated so that the larger its value, the larger the prediction error.
• the degree of anomaly of the explanatory variable x * of the sample data compared with the reference data can be used as the index s * x for the explanatory variable of the sample data.
• the Mahalanobis distance of the explanatory variable x * of the sample data with respect to the reference data D is used as the index s * x for the degree of anomaly of the explanatory variable x * of the sample data.
• the index s * x can be expressed by the following formula 1.
• the degree of anomaly of the objective variable y * of the sample data compared with the reference data can be used as the index s * y for the objective variable of the sample data.
• the Mahalanobis distance of the objective variable y * of the sample data with respect to the reference data D is used as the index s * y for the degree of anomaly of the objective variable y * of the sample data.
• the index s * y can be expressed by the following formula 2.
• the index s * y for the objective variable of the sample data can be the variance representing the degree of variation of the objective variable of the reference data D related to the sample data, that is, the reference data D located in the vicinity of the objective variable y * of the sample data.
• the variance representing the degree of variation of the objective variable of the reference data D related to the sample data, that is, the reference data D located in the vicinity of the objective variable y * of the sample data.
• the performance evaluation value of the prediction model using the reference data D related to the sample data can be used.
• the index s * f can be expressed as the mean squared error (MSE) for the prediction model f of the reference data D located in the vicinity of the sample data, as shown in the following formula 3.
• MSE mean squared error
• the prediction error L * can be expressed as the sum of the contributions corresponding to the indexes, and can be expressed by the following formula 4, for example.
• the above L * 0 is a contribution from other than the calculated index (for example, offset or unknown error), and can be equally divided and distributed into the contribution amount corresponding to each index.
• the error decomposition unit 15 uses a contribution calculation function that decomposes the prediction error L * into each contribution amount L * i at the ratio of each index s * .
• the contribution calculation function is expressed by the following formula 5, and thus can be decomposed into the contribution amount L * i of each index as shown in formula 6.
• the error decomposition unit 15 is not necessarily limited to calculating each contribution amount corresponding to each index using the method described above.
• the error decomposition unit 15 may decompose the prediction error into each contribution amount using another method such as that described in the third embodiment below.
• the information processing device 10 acquires a prediction model f that has been generated in advance by machine learning, reference data D that can be input to the prediction model f, and sample data that is the subject of decomposition of prediction error (step S1).
• the information processing device 10 calculates the prediction error L * when the sample data is input to the prediction model f (step S2). For example, the information processing device 10 calculates the square error (y*-f(x * )) 2 between the predicted value f(x * ), which is the output when the explanatory variable x * of the sample data is input to the prediction model f , and the objective variable y * of the sample data, as the prediction error L * .
• the information processing device 10 also calculates an index s * for evaluating the contribution of each data to the prediction error (step S3).
• the index s* is calculated as an index s * , which is an index s * x for the explanatory variable of the sample data, an index s * y for the objective variable of the sample data, and an index s * f for the prediction model.
• the index s* may be an anomaly level of the explanatory variable x * of the sample data compared to the reference data, an anomaly level of the objective variable y * of the sample data compared to the reference data, a variance representing the degree of variation of the objective variable of the reference data D located in the vicinity of the objective variable y * of the sample data, or a performance evaluation value of the prediction model using the reference data D located in the vicinity of the sample data.
• the information processing device 10 calculates the contribution amount L * i of each of the explanatory variable x* of the sample data, the objective variable y* of the sample data, and the prediction model f in the prediction error L * using the prediction error L* and the index s * calculated as described above (step S4).
• the contribution amount L * i corresponding to each index is calculated using a contribution calculation function that breaks down the prediction error L * into each contribution amount L * i based on the ratio of each index s * .
• the information processing device 10 outputs the calculated contribution amount L * i corresponding to each index by displaying it on a display device (step S5). For example, as shown in FIG. 1, the information processing device 10 outputs the contribution amount of the explanatory variable of the sample data, the contribution amount of the target variable of the sample data, and the contribution amount of the prediction model to the prediction error.
• each index for evaluating the contribution of each piece of data to the prediction error is calculated based on the data used to calculate the prediction error of the prediction error including the data used in the prediction model f, and the contribution of each piece of data to the prediction error is calculated based on these indexes.
• an index for the explanatory variable of the sample data, an index for the objective variable of the sample data, and an index for the prediction model are calculated, and these are decomposed and calculated into their respective contributions to the prediction error.
• the index calculation unit 14 in this embodiment generates each index using a check model g (second prediction model), which is a different prediction model from the prediction model f and is generated using the above-mentioned prediction model f or the reference data D.
• the check model g is a model generated by machine learning separately to evaluate the performance of the prediction model f, and is one or more models.
• the check model g is, for example, trained with different hyperparameters using the same learning algorithm as the prediction model f, trained with a dataset included in the reference data D that is different from the dataset used to train the prediction model f using the same learning algorithm as the prediction model f, or trained with the reference data D (training data, etc.) using a learning algorithm different from that of the prediction model f.
• the index calculation unit 14 regards the check model g as a true model, and generates a plurality of indexes s * using the output g(x * ) obtained by inputting the explanatory variable x * of the sample data to the check model g. Specifically, the index calculation unit 14 calculates the index using the variance V and expected value E of the output by the check model g calculated using m check models g, as shown in the following formula 7.
• the index calculation unit 14 calculates the index s*x for the explanatory variable x * of the reference data, the index s * x for the objective variable x * of the reference data, and the index s* f for the prediction model f, as shown in the following formula 8 , and sets each index s * x , s * y , and s * f as the contribution amount L * x , L * y , and L * f corresponding to each index as it is.
• L * 0 is an offset or other unknown error.
• the error decomposition unit 15 sets the contribution calculation function by the identity that the sum of the above-mentioned indices s * x , s * y , s * f, that is, the contribution amounts L * x , L * y , L * f and other contribution amounts L * 0 is equal to the prediction error L * of the prediction model f.
• the prediction error L * is a square error
• the identity of the following formula 9 is established, so that the contribution calculation function that makes the calculated index the contribution to the error as it is can be used.
• the error decomposition unit 15 can use such a contribution calculation function to calculate the contribution amounts L * x , L * y , L * f corresponding to the indices s * x , s * y , s * f .
• the indices for evaluating the contribution of each piece of data to the prediction error are set using the check model g, and the contribution of each piece of data to the prediction error is calculated based on these indices.
• Fig. 5 is a diagram for explaining the configuration of an information processing device 10
• Fig. 6 is a diagram for explaining processing by the information processing device 10.
• the information processing device 10 in this embodiment includes a contribution calculation function learning unit 18 and an error regression model storage unit 19 in addition to the configuration of the information processing device 10 described in the above-mentioned embodiments 1 and 2.
• the function of the contribution calculation function learning unit 18 can be realized by the calculation device executing a program for realizing each function stored in the storage device.
• the error regression model storage unit 19 is composed of a storage device. Each component will be described in detail below. Below, the components that are different from the above-mentioned embodiment 1 or embodiment 2 will be mainly described in detail.
• the error L is the error L between the predicted value f(x), which is the output when the explanatory variable x of the reference data D, such as training data, is input to the prediction model f, and the objective variable y of the reference data D.
• Such error L and index s can be calculated, for example, by using the prediction model f and reference data D using the function of the index calculation unit 14 described above.
• the contribution calculation function learning unit 18 learns the index s generated from the prediction model f and the reference data D as described above as an explanatory variable and the error L as a target variable, and generates an error regression model h(s).
• the contribution calculation function learning unit 18 stores the generated error regression model h(s) in the error regression model storage unit 19.
• the contribution calculation function learning unit 18 may select and learn the index s using a feature selection method. For example, different indexes s may be combined and learned, and the index s may be selected so as to improve the performance of the error regression model h(s).
• the error decomposition unit 15 (contribution calculation unit) in this embodiment generates a contribution calculation function using the above-mentioned error regression model h(s).
• the error decomposition unit 15 when the error regression model h(s) is a linear model, the error decomposition unit 15 generates a contribution calculation function using the weight parameter w i set in the error regression model by learning.
• the contribution calculation function is generated so that the value obtained by multiplying each weight parameter w i assigned to each index s x , s y , and s f in the error regression model by each index s * x , s * y , and s * f calculated in the first embodiment as described above is the contribution amount L * l corresponding to each index s * l .
• the contribution calculation function according to the following formula 10 can be generated and decomposed into the contribution amount L * l corresponding to each index using the above-mentioned prediction error L * as shown in formula 11.
• the error decomposition unit 15 interprets the output when the indexes s * x , s * y , and s * f calculated in the first embodiment are input to the error regression model h(s) instead of the indexes sx , sy , and sf as described above, and generates a contribution calculation function.
• the output of the error regression model can be calculated by using a model interpretation method that can express the contribution of each index as the sum of the contributions of each index.
• a contribution calculation function according to the following Formula 13 is generated.
• v * l is the contribution from each index, and is, for example, the SHAP value (SHAPley Additive explanations values) of s * l .
• the prediction error L * described above can be used to decompose into the contribution amount L * l corresponding to each index.
• a model that learns the relationship between the index and the error is generated, and the contribution of the feature value s i to the error of each data point with respect to the prediction error is calculated based on the model.
• Fig. 7 to Fig. 8 are block diagrams showing the configuration of an information processing device in the fourth embodiment. Note that this embodiment shows an outline of the configuration of the information processing device described in the above-mentioned embodiment.
• the information processing device 100 is configured as a general information processing device, and is equipped with the following hardware configuration, as an example.
• ⁇ CPU Central Processing Unit
• ROM Read Only Memory
• RAM Random Access Memory
• Program group 104 loaded into RAM 103
• a storage device 105 for storing the program group 104
• a drive device 106 that reads and writes data from and to a storage medium 110 outside the information processing device.
• a communication interface 107 that connects to a communication network 111 outside the information processing device
• Input/output interface 108 for inputting and outputting data
• a bus 109 that connects each component
• FIG. 7 shows an example of the hardware configuration of the information processing device 100, and the hardware configuration of the information processing device is not limited to the above-mentioned case.
• the information processing device may be configured with a part of the above-mentioned configuration, such as not having the drive device 106.
• the information processing device may use a GPU (Graphic Processing Unit), a DSP (Digital Signal Processor), an MPU (Micro Processing Unit), an FPU (Floating point number Processing Unit), a PPU (Physics Processing Unit), a TPU (Tensor Processing Unit), a quantum processor, a microcontroller, or a combination of these.
• the information processing device 100 can be equipped with the error calculation unit 121, index calculation unit 122, and contribution calculation unit 123 shown in FIG. 8 by having the CPU 101 acquire and execute the program group 104.
• the program group 104 is stored in the storage device 105 or ROM 102 in advance, for example, and the CPU 101 loads the program group 104 into the RAM 103 and executes it as necessary.
• the program group 104 may be supplied to the CPU 101 via the communication network 111, or may be stored in the storage medium 110 in advance, and the drive device 106 may read out the program and supply it to the CPU 101.
• the error calculation unit 121, index calculation unit 122, and contribution calculation unit 123 described above may be constructed with dedicated electronic circuits for realizing such means.
• the error calculation unit 121 calculates the prediction error between the predicted value, which is the output when the explanatory variables of the target data are input to the prediction model, and the objective variable of the target data.
• the index calculation unit 122 calculates an index for evaluating the contribution to the prediction error in at least one of the explanatory variables of the target data, the objective variable of the target data, and the prediction model, based on the data used to calculate the prediction error. For example, the index calculation unit 122 calculates the index using data on at least one of the explanatory variables of the target data, the objective variable of the target data, and the prediction model, and reference data used when generating the prediction model.
• the contribution calculation unit 123 calculates the contribution amount based on the prediction error and the index.
• Non-transitory computer readable medium includes various types of tangible storage medium.
• Examples of non-transitory computer readable medium include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROM (Read Only Memory), CD-R, CD-R/W, and semiconductor memory (e.g., mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (Random Access Memory)).
• the program may also be supplied to a computer by various types of transitory computer readable medium. Examples of transitory computer readable medium include electrical signals, optical signals, and electromagnetic waves.
• the temporary computer-readable medium can provide the program to the computer via a wired communication path, such as an electric wire or optical fiber, or via a wireless communication path.
• (Appendix 1) an error calculation unit that calculates a prediction error between a predicted value, which is an output when an explanatory variable of the target data is input to the prediction model, and a response variable of the target data; an index calculation unit that calculates an index for evaluating a contribution to the prediction error in at least one of the explanatory variables of the target data, the objective variable of the target data, and the prediction model, based on data that can be used to calculate the prediction error; a contribution calculation unit that calculates the contribution amount based on the prediction error and the index;
• An information processing device comprising: (Appendix 2) 2.
• the information processing device calculates the index using at least one data of the explanatory variables of the target data, the objective variables of the target data, and the prediction model, and reference data used in the prediction model.
• Information processing device. (Appendix 3) 3.
• the information processing device according to claim 2 The index calculation unit calculates the index for at least one of the explanatory variables and the objective variable of the target data by using the target data and the reference data.
• Information processing device. (Appendix 4) 4.
• the information processing device according to claim 3 The index calculation unit calculates the index for at least one of the explanatory variables and the objective variable of the target data based on a comparison result between the target data and the reference data.
• Information processing device. (Appendix 5) 4.
• the index calculation unit calculates the index for the objective variable of the target data based on a degree of variation of the objective variable of the reference data related to the target data.
• Information processing device. An information processing device according to any one of Supplementary Note 2 to 5, The index calculation unit calculates the index in the prediction model based on a performance evaluation value of the prediction model calculated using the reference data related to the target data.
• Information processing device. (Appendix 7) 7. The information processing device according to claim 1, The index calculation unit generates the index based on at least the prediction model or a second prediction model generated based on reference data used in the prediction model and the target data. Information processing device. (Appendix 8) 8.
• the information processing device calculates a plurality of the indexes based on an output of the second prediction model in which an explanatory variable of the target data is input; the contribution calculation unit calculates the contribution amount based on values of a plurality of the indices, the sum of which is a value based on the prediction error; Information processing device. (Appendix 9) 9.
• the index calculation unit calculates the indexes for at least the explanatory variables of the target data, the objective variable of the target data, and the prediction model based on an output of an explanatory variable of the target data input to the second prediction model; the contribution calculation unit calculates the contribution based on a value of the index such that a value including a sum of all of the indexes and a value based on the prediction error are equal to each other; Information processing device. (Appendix 10) 10.
• the information processing device configured to learn a model that represents a relationship between an error between an output when an explanatory variable of reference data used in the prediction model is input to the prediction model and a response variable of the reference data, and a second index for evaluating a contribution to the error in each of the explanatory variable of the reference data, the response variable of the reference data, and the prediction model; the contribution calculation unit calculates the contribution amount based on the model, the index, and the prediction error; Information processing device. (Appendix 11) 11. The information processing device according to claim 10, the contribution calculation unit calculates the contribution amount based on a contribution degree of the index to an output when the index is input to the model; Information processing device. (Appendix 12) 12.
• the information processing device selects the second index to learn the model.
• Information processing device (Appendix 13) Calculating a prediction error between a predicted value, which is an output when an explanatory variable of the target data is input to a prediction model, and a response variable of the target data; Calculating an index for evaluating a contribution to the prediction error in at least one of the explanatory variables of the target data, the objective variable of the target data, and the prediction model, based on data that can be used to calculate the prediction error; Calculating the contribution amount based on the prediction error and the index.
• Information processing methods (Appendix 14) 14.
• the information processing method generating the index based on the target data and a second prediction model generated based on at least the prediction model or the reference data used in the prediction model; Information processing methods.
• (Appendix 15) 15 The information processing method according to claim 13, further comprising: learning a model that represents a relationship between an error between an output when an explanatory variable of the reference data is input to the prediction model and a response variable of the reference data, and a second index for evaluating a contribution to the error of each of the explanatory variable of the reference data, the response variable of the reference data, and the prediction model; calculating the contribution based on the model, the indicators, and the prediction error; Information processing methods.
• (Appendix 16) Calculating a prediction error between a predicted value, which is an output when an explanatory variable of the target data is input to a prediction model, and a response variable of the target data; Calculating an index for evaluating a contribution to the prediction error in at least one of the explanatory variables of the target data, the objective variable of the target data, and the prediction model, based on data that can be used to calculate the prediction error; Calculating the contribution amount based on the prediction error and the index.
• a computer-readable storage medium that stores a program for causing a computer to execute a process.
• Information processing device 11 Decomposition target sample input unit 12 Decomposition result output unit 13 Error decomposition unit 14 Index calculation unit 15 Error decomposition unit 16 Prediction model storage unit 17 Reference data storage unit 18 Contribution calculation function learning unit 19 Error regression model storage unit 100 Information processing device 101 CPU 102 ROM 103 RAM Reference Signs List 104 Program Group 105 Storage Device 106 Drive Device 107 Communication Interface 108 Input/Output Interface 109 Bus 110 Storage Medium 111 Communication Network 121 Error Calculation Unit 122 Index Calculation Unit 123 Contribution Calculation Unit

Landscapes

• Engineering & Computer Science (AREA)
• Theoretical Computer Science (AREA)
• General Engineering & Computer Science (AREA)
• General Physics & Mathematics (AREA)
• Physics & Mathematics (AREA)
• Evolutionary Computation (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Software Systems (AREA)
• Data Mining & Analysis (AREA)
• Artificial Intelligence (AREA)
• Quality & Reliability (AREA)
• Health & Medical Sciences (AREA)
• Biomedical Technology (AREA)
• Bioinformatics & Cheminformatics (AREA)
• Bioinformatics & Computational Biology (AREA)
• Life Sciences & Earth Sciences (AREA)
• Evolutionary Biology (AREA)
• Medical Informatics (AREA)
• Computing Systems (AREA)
• Mathematical Physics (AREA)
• Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Priority Applications (4)

Applications Claiming Priority (1)

Publications (1)

Family

ID=92589480

Family Applications (2)

Family Applications After (1)

Country Status (3)

Cited By (1)

Citations (3)

• 2023
  • 2023-02-28 WO PCT/JP2023/007228 patent/WO2024180648A1/ja not_active Ceased
  • 2023-08-23 JP JP2025503567A patent/JPWO2024180789A1/ja active Pending
  • 2023-08-23 WO PCT/JP2023/030357 patent/WO2024180789A1/ja not_active Ceased
  • 2023-08-23 US US18/578,803 patent/US20250103422A1/en active Pending

Patent Citations (3)

Also Published As

Similar Documents

Legal Events