WO2024154230A1 - 情報処理装置、情報処理方法、及び記録媒体 - Google Patents
情報処理装置、情報処理方法、及び記録媒体 Download PDFInfo
- Publication number
- WO2024154230A1 WO2024154230A1 PCT/JP2023/001187 JP2023001187W WO2024154230A1 WO 2024154230 A1 WO2024154230 A1 WO 2024154230A1 JP 2023001187 W JP2023001187 W JP 2023001187W WO 2024154230 A1 WO2024154230 A1 WO 2024154230A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- likelihood ratio
- information processing
- score
- processing device
- sequence data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Definitions
- This disclosure relates to the technical fields of information processing devices, information processing methods, and recording media.
- Patent Document 1 discloses classifying sequence data into one of a number of predefined classes by sequentially acquiring and analyzing multiple elements contained in the sequence data.
- Patent Document 2 discloses determining a threshold value for the score each time multiple elements are acquired.
- This disclosure aims to improve the related technology described above.
- One aspect of the information processing device disclosed herein includes an acquisition means for acquiring multiple elements included in sequence data, a likelihood ratio calculation means for inputting the multiple elements and calculating the relationship between each element to calculate a likelihood ratio indicating the likelihood of the class to which the sequence data belongs, a score calculation means for adding a value corresponding to an urgency signal to the likelihood ratio to calculate a score, and a classification means for classifying the sequence data into at least one of multiple classes that are classification candidates based on the score.
- One aspect of the information processing method disclosed herein involves at least one computer acquiring multiple elements contained in sequence data, inputting the multiple elements and calculating the relationships between the elements to calculate a likelihood ratio indicating the likelihood of the class to which the sequence data belongs, adding a value corresponding to an urgency signal to the likelihood ratio to calculate a score, and classifying the sequence data into at least one of multiple candidate classes based on the score.
- a computer program is recorded on at least one computer to execute an information processing method that acquires multiple elements contained in sequence data, inputs the multiple elements and calculates the relationships between the elements to calculate a likelihood ratio indicating the likelihood of the class to which the sequence data belongs, calculates a score by adding a value corresponding to an urgency signal to the likelihood ratio, and classifies the sequence data into at least one of multiple classes that are classification candidates based on the score.
- FIG. 1 is a block diagram showing a hardware configuration of an information processing device according to a first embodiment.
- 1 is a block diagram showing a functional configuration of an information processing device according to a first embodiment.
- 5 is a flowchart showing a flow of operations of the information processing device according to the first embodiment.
- 11 is a graph (part 1) showing an example of a score calculated by the information processing device according to the first embodiment together with a comparative example.
- 11 is a graph (part 2) showing an example of a score calculated by the information processing device according to the first embodiment together with a comparative example.
- FIG. 11 is a block diagram showing a functional configuration of an information processing device according to a second embodiment.
- 10 is a flowchart showing a flow of operations of the information processing device according to the second embodiment.
- FIG. 13 is a block diagram showing a functional configuration of an information processing device according to a third embodiment. 13 is a flowchart showing the flow of operations of the information processing device according to the third embodiment.
- FIG. 13 is a block diagram showing a functional configuration of an information processing device according to a fourth embodiment. 13 is a flowchart showing the flow of operations of the information processing device according to the fourth embodiment.
- FIG. 13 is a block diagram showing a functional configuration of an information processing device according to a fifth embodiment. 13 is a flowchart showing a flow of a likelihood ratio calculation operation in the information processing device according to the fifth embodiment.
- FIG. 1 An information processing apparatus according to a first embodiment will be described with reference to FIGS. 1 to 5.
- FIG. 1 An information processing apparatus according to a first embodiment will be described with reference to FIGS. 1 to 5.
- FIG. 1 An information processing apparatus according to a first embodiment will be described with reference to FIGS. 1 to 5.
- Fig. 1 is a block diagram showing the hardware configuration of the information processing apparatus according to the first embodiment.
- the information processing device 10 includes a processor 11, a RAM (Random Access Memory) 12, a ROM (Read Only Memory) 13, and a storage device 14.
- the information processing device 10 may further include an input device 15 and an output device 16.
- the above-mentioned processor 11, RAM 12, ROM 13, storage device 14, input device 15, and output device 16 are each connected via a data bus 17.
- the processor 11 reads a computer program.
- the processor 11 is configured to read a computer program stored in at least one of the RAM 12, the ROM 13, and the storage device 14.
- the processor 11 may read a computer program stored in a computer-readable storage medium using a storage medium reading device (not shown).
- the processor 11 may obtain (i.e., read) a computer program from a device (not shown) disposed outside the information processing device 10 via a network interface.
- the processor 11 controls the RAM 12, the storage device 14, the input device 15, and the output device 16 by executing the computer program that the processor 11 reads.
- a functional block that performs class classification based on a likelihood ratio is realized within the processor 11.
- the processor 11 may function as a controller that performs each control in the information processing device 10.
- the processor 11 may be configured as, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (field-programmable gate array), a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), or a quantum processor.
- the processor 11 may be configured as one of these, or may be configured to use multiple processors in parallel.
- RAM 12 temporarily stores computer programs executed by processor 11.
- RAM 12 temporarily stores data that processor 11 uses temporarily while processor 11 is executing a computer program.
- RAM 12 may be, for example, a D-RAM (Dynamic Random Access Memory) or an SRAM (Static Random Access Memory). Also, other types of volatile memory may be used instead of RAM 12.
- ROM 13 stores computer programs executed by processor 11. ROM 13 may also store other fixed data. ROM 13 may be, for example, a P-ROM (Programmable Read Only Memory) or an EPROM (Erasable Read Only Memory). Also, other types of non-volatile memory may be used instead of ROM 13.
- the storage device 14 stores data that the information processing device 10 stores long-term.
- the storage device 14 may operate as a temporary storage device for the processor 11.
- the storage device 14 may include, for example, at least one of a hard disk device, a magneto-optical disk device, an SSD (Solid State Drive), and a disk array device.
- the input device 15 is a device that receives input instructions from a user of the information processing device 10.
- the input device 15 may include, for example, at least one of a keyboard, a mouse, and a touch panel.
- the input device 15 may be configured as a mobile terminal such as a smartphone or a tablet.
- the input device 15 may be, for example, a device that includes a microphone and is capable of voice input.
- the output device 16 is a device that outputs information related to the information processing device 10 to the outside.
- the output device 16 may be a display device (e.g., a display) that can display information related to the information processing device 10.
- the output device 16 may also be a speaker or the like that can output information related to the information processing device 10 as audio.
- the output device 16 may be configured as a mobile terminal such as a smartphone or a tablet.
- FIG. 1 shows an example of an information processing device 10 that is configured to include multiple devices, all or some of these functions may be realized by a single device.
- Such an information processing device may, for example, be configured to include only the above-mentioned processor 11, RAM 12, and ROM 13, and the other components (i.e., storage device 14, input device 15, output device 16, etc.) may be provided by an external device connected to the information processing device 10.
- the information processing device 10 may have some of its calculation functions realized by an external device (for example, an external server or cloud, etc.).
- Fig. 2 is a block diagram showing the functional configuration of the information processing device according to the first embodiment.
- the information processing device 10 is a device that performs class classification of input sequence data, and is configured to include a data acquisition unit 50, a likelihood ratio calculation unit 100, a score calculation unit 150, and a class classification unit 200 as components for realizing the function.
- Each of the data acquisition unit 50, the likelihood ratio calculation unit 100, the score calculation unit 150, and the class classification unit 200 may be a processing block realized by, for example, the above-mentioned processor 11 (see FIG. 1).
- the data acquisition unit 50 is configured to be able to acquire multiple elements contained in the sequence data.
- the data acquisition unit 50 may acquire data directly from any data acquisition device (e.g., a camera, a microphone, etc.), or may read data that has been acquired in advance by a data acquisition device and stored in storage, etc.
- the data acquisition unit 50 may be configured to acquire data from each of multiple cameras.
- the elements of the sequence data acquired by the data acquisition unit 50 are configured to be output to the likelihood ratio calculation unit 100.
- sequence data is data that includes multiple elements arranged in a predetermined order, and one example is time-series data. More specific examples of sequence data include, but are not limited to, video data, audio data, or subdivided image data.
- the likelihood ratio calculation unit 100 is configured to be able to calculate a likelihood ratio based on the multiple elements acquired by the data acquisition unit 50.
- the likelihood ratio calculation unit 100 may be configured to be able to calculate a likelihood ratio based on the relationship between at least two consecutive elements among the multiple elements acquired.
- the "likelihood ratio" here is an index that indicates the likelihood of the class to which the sequence data belongs. Specific examples of likelihood ratios and specific calculation methods will be described in detail in other embodiments described later.
- the likelihood ratio calculated by the likelihood ratio calculation unit 100 is transmitted to the score calculation unit 150.
- the score calculation unit 150 is configured to be able to calculate a score for classifying based on the likelihood ratio calculated by the likelihood ratio calculation unit 100. Specifically, the score calculation unit 150 is configured to be able to calculate a score by adding a value corresponding to the urgency signal to the likelihood ratio.
- the "urgency signal” here is a signal that indicates the urgency of a response, and is used, for example, when making an early decision (see, for example, References 1 and 2 below).
- the score calculation unit 150 adds the above-mentioned urgency signal to the likelihood ratio to calculate a score that is greater than when the score is simply calculated from the likelihood ratio. That is, the urgency signal according to this embodiment is used to increase the calculated score.
- the score calculation unit 150 may calculate the score, for example, by calculating the sum of the likelihood ratio and the urgency signal. Alternatively, the score calculation unit 150 may calculate the score by multiplying the likelihood ratio by a coefficient corresponding to the urgency signal. That is, the process of "adding" the urgency signal according to this embodiment may be a calculation process that uses the urgency signal to increase the score calculated from the likelihood ratio.
- the value corresponding to the urgency signal may be calculated using, for example, a fixed function form.
- the function form in this case is not particularly limited, and may be, for example, a constant, linear, or non-linear.
- the value corresponding to the urgency signal may be input from outside the information processing device 10, or may be stored inside the information processing device 10. Alternatively, the value corresponding to the urgency signal may be generated appropriately in the information processing device 10. The configuration for generating the value corresponding to the urgency signal will be described in detail in other embodiments described later.
- the classifying unit 200 is configured to classify the sequence data based on the score calculated by the score calculation unit 150 (i.e., the score obtained by adding the urgency signal to the likelihood ratio).
- the classifying unit 200 selects at least one class to which the sequence data belongs from among a plurality of classes that are classification candidates.
- the plurality of classes that are classification candidates may be preset. Alternatively, the plurality of classes that are classification candidates may be appropriately set by the user, or may be appropriately set based on the type of sequence data to be handled.
- the plurality of classes may be set as a class indicating that the face included in the video data is a real face (i.e., a biological face) and a class indicating that the face is a fake face (e.g., a face in a photograph or a 3D mask).
- the information processing device 10 can be used as a spoofing detection device.
- the plurality of classes may be set as a class indicating that an object included in a video taken in a moving body such as a car is an obstacle to be avoided and a class indicating that the object is an object other than an obstacle.
- the information processing device 10 can be used as an obstacle detection device.
- the number of classes that are classification candidates is not limited to two, but may be three or more classes.
- Fig. 3 is a flowchart showing the flow of operations of the information processing device according to the first embodiment.
- the data acquisition unit 50 first acquires elements contained in the sequence data (step S11).
- the data acquisition unit 50 outputs the acquired elements of the sequence data to the likelihood ratio calculation unit 100.
- the likelihood ratio calculation unit 100 then calculates a likelihood ratio from the relationship between the acquired multiple elements (step S12).
- the score calculation unit 150 adds the urgency signal to the likelihood ratio calculated by the likelihood ratio calculation unit 100 to calculate a score to be used for class classification (step S13). Then, the class classification unit 200 performs class classification based on the calculated score (step S14).
- the class classification may determine one class to which the sequence data belongs, or may determine multiple classes to which the sequence data is likely to belong.
- the class classification unit 200 may have a function of outputting the result of the class classification.
- the class classification unit 200 may output the result of the class classification to a display or the like.
- the class classification unit 200 may output the result of the class classification as sound via a speaker or the like.
- the class classification unit 200 may calculate the likelihood ratio again without performing class classification (i.e., without determining the class to classify into).
- the data acquisition unit 50 may acquire new elements contained in the sequence data, and a new likelihood ratio may be calculated.
- Fig. 4 is a graph (part 1) showing an example of a score calculated by the information processing device according to the first embodiment together with a comparative example.
- Fig. 5 is a graph (part 2) showing an example of a score calculated by the information processing device according to the first embodiment together with a comparative example.
- the class classification unit 200 may determine a class to classify when the calculated score reaches a predetermined judgment threshold.
- an urgency signal is added to the likelihood ratio to calculate the score, so the score increases relatively rapidly. Therefore, the score calculated in this embodiment reaches the judgment threshold at a relatively early timing (time T1 in the figure).
- the score is calculated without adding an urgency signal to the likelihood ratio. Therefore, the score increases relatively slowly. Therefore, the score calculated in the comparative example reaches the judgment threshold at a later timing (time T2 in the figure) than in this embodiment. In this way, according to the information processing device 10 according to the present embodiment, early class classification can be realized using the score to which the urgency signal is added.
- the information processing device 10 may set a maximum time Tmax allowed for classifying. For example, when classifying using a video of a person passing in front of a camera, it is required to finish classifying before the person moves out of the camera's field of view. In such a case, in a comparative example in which an urgency signal is not added to the likelihood ratio, the score may not reach the judgment threshold by the maximum time Tmax. In other words, the information processing device 10 according to the comparative example may not be able to perform classifying normally. However, in this embodiment, the score is calculated by adding an urgency signal to the likelihood ratio, so that the score reaches the judgment threshold at time T3 before the maximum time Tmax. In this way, the information processing device 10 according to this embodiment can appropriately classify sequence data even when there is a limit to the time allowed for classifying.
- a method of varying the judgment threshold according to the elapsed time i.e., using a dynamic threshold
- rigorous calculations are required (for example, calculating the likelihood ratio up to the maximum time Tmax and then calculating backwards).
- the calculations for setting the optimal dynamic threshold take time, and as a result, it takes longer to perform classification.
- the information processing device 10 according to the second embodiment will be described with reference to Fig. 6 and Fig. 7.
- the second embodiment differs from the first embodiment only in some configurations and operations, and other parts may be the same as the first embodiment. Therefore, the following will describe in detail the parts that differ from the first embodiment, and will omit descriptions of other overlapping parts as appropriate.
- Fig. 6 is a block diagram showing the functional configuration of the information processing device according to the second embodiment.
- the same reference numerals are used to designate the same elements as those shown in Fig. 2.
- the information processing device 10 according to the second embodiment is configured to include, as components for realizing its functions, a data acquisition unit 50, a likelihood ratio calculation unit 100, a score calculation unit 150, a class classification unit 200, and an urgency determination unit 310. That is, the information processing device 10 according to the second embodiment further includes an urgency determination unit 310 in addition to the configuration of the first embodiment (see FIG. 2).
- the urgency determination unit 310 may be a processing block realized by, for example, the above-mentioned processor 11 (see FIG. 1).
- the urgency determination unit 310 is configured to be able to determine whether or not the current state is an emergency state.
- an "emergency state” refers to a state in which it is possible that the sequence data cannot be classified (i.e., the score does not reach the determination threshold) before the maximum time Tmax has elapsed. Whether or not an emergency state exists may be determined, for example, using the likelihood ratio calculated by the likelihood ratio calculation unit 100. For example, the urgency determination unit 310 may determine whether or not an emergency state exists using the value of the likelihood ratio or the rate of change.
- the determination result by the urgency determination unit 310 is configured to be output to the score calculation unit 150.
- the score calculation unit 150 according to the second embodiment calculates a score according to the determination result of the urgency determination unit 310. The operation of the score calculation unit 150 at this time will be described in detail below.
- FIG. 7 is a flowchart showing the flow of operations of the information processing device according to the second embodiment. Note that in Fig. 7, the same processes as those shown in Fig. 3 are denoted by the same reference numerals.
- the data acquisition unit 50 first acquires elements contained in the sequence data (step S11).
- the data acquisition unit 50 outputs the acquired elements of the sequence data to the likelihood ratio calculation unit 100.
- the likelihood ratio calculation unit 100 then calculates a likelihood ratio from the relationship between the acquired multiple elements (step S12).
- the urgency determination unit 310 determines whether or not there is an emergency state based on the calculated likelihood ratio, etc. (step S15). If it is determined that there is an emergency state (step S15: YES), the score calculation unit 150 adds an urgency signal to the likelihood ratio calculated by the likelihood ratio calculation unit 100 to calculate a score to be used for class classification (step S13). On the other hand, if it is determined that there is no emergency state (step S15: NO), the score calculation unit 150 calculates a score to be used for class classification from the likelihood ratio calculated by the likelihood ratio calculation unit 100 (without adding an urgency signal) (step S16).
- the classifying unit 200 performs classifying based on the calculated score (step S14). That is, if it is an emergency state, the classifying unit 200 performs classifying using the score to which the urgency signal has been added. On the other hand, if it is not an emergency state, the classifying unit 200 performs classifying using the score to which the urgency signal has not been added. Note that if it is not determined to be an emergency state at the initial stage, but is determined to be an emergency state halfway through, the score to which the urgency signal has not been added until halfway through may be used, and the score to which the urgency signal has been added after it has been determined to be an emergency state may be used.
- the information processing device 10 if it is determined that an emergency state exists, a value corresponding to the urgency signal is added to calculate the score, and if it is determined that there is no emergency state, the score is calculated without adding a value corresponding to the urgency. In this way, it is possible to prevent the urgency signal from being added unnecessarily, and more appropriate class classification can be achieved.
- the information processing device 10 according to the third embodiment will be described with reference to Figures 8 and 9.
- the third embodiment differs from the first and second embodiments only in some configurations and operations, and other parts may be the same as the first embodiment. Therefore, hereinafter, the parts that differ from the embodiments already described will be described in detail, and the explanation of other overlapping parts will be omitted as appropriate.
- Fig. 8 is a block diagram showing the functional configuration of the information processing device according to the third embodiment.
- the same elements as those shown in Fig. 2 are denoted by the same reference numerals.
- the information processing device 10 according to the third embodiment is configured to include, as components for realizing its functions, a data acquisition unit 50, a likelihood ratio calculation unit 100, a score calculation unit 150, a class classification unit 200, and an urgency signal generation unit 320. That is, the information processing device 10 according to the third embodiment further includes an urgency signal generation unit 320 in addition to the configuration of the first embodiment (see FIG. 2).
- the urgency signal generation unit 320 may be a processing block realized by, for example, the above-mentioned processor 11 (see FIG. 1).
- the urgency signal generating unit 320 is configured to be able to generate a value corresponding to the urgency signal using a trained model. That is, the urgency signal generating unit 320 is configured to be able to generate a value corresponding to the urgency that is used by the score calculation unit 150 when calculating the score.
- the trained model may include, for example, a neural network.
- the trained model may be trained as a model that receives, as input, a value such as a likelihood ratio, and outputs a value corresponding to the urgency signal.
- the trained model may be trained by machine learning using training data prepared in advance.
- the trained model may be trained using a loss function plus a value corresponding to the time that has elapsed since the acquisition of multiple elements began. More specifically, the trained model may be trained using the loss function LOSS shown in the following formula (1).
- t here is the current time (i.e., the time that has elapsed since the start of element acquisition).
- Const is a predetermined constant, and may be a value that is appropriately set by a user, etc., depending on how quickly the classification needs to be performed.
- Fig. 9 is a flowchart showing the flow of operations of the information processing device according to the third embodiment. Note that in Fig. 9, the same processes as those shown in Fig. 3 are denoted by the same reference numerals.
- the data acquisition unit 50 first acquires elements contained in the sequence data (step S11).
- the data acquisition unit 50 outputs the acquired elements of the sequence data to the likelihood ratio calculation unit 100.
- the likelihood ratio calculation unit 100 then calculates a likelihood ratio from the relationship between the acquired multiple elements (step S12).
- the urgency signal generation unit 150 generates a value corresponding to the urgency signal using the trained model (step S17). Then, the score calculation unit 150 adds the value generated by the emergency mental synthesis unit 150 to the likelihood ratio calculated by the likelihood ratio calculation unit 100 to calculate a score to be used for class classification (step S13).
- the classifying unit 200 performs classifying based on the calculated score (step S14).
- the classifying may determine one class to which the sequence data belongs, or may determine multiple classes to which the sequence data is likely to belong.
- a value corresponding to the urgency signal is generated using a trained model.
- the value generated in this manner it is possible to calculate the score more appropriately, and it is possible to realize early class classification.
- the information processing device 10 according to the fourth embodiment will be described with reference to Fig. 10 and Fig. 11.
- the fourth embodiment differs from the first to third embodiments in part of the configuration and operation, and other parts may be the same as the first to third embodiments. Therefore, hereinafter, the parts that differ from the embodiments already described will be described in detail, and the explanation of other overlapping parts will be omitted as appropriate.
- Fig. 10 is a block diagram showing the functional configuration of the information processing device according to the fourth embodiment.
- the same reference numerals are used to designate the same elements as those shown in Fig. 2.
- the information processing device 10 is configured to include, as components for realizing its functions, a data acquisition unit 50, a likelihood ratio calculation unit 100, a score calculation unit 150, and a class classification unit 200.
- the likelihood ratio calculation unit 100 includes a first calculation unit 110 and a second calculation unit 120.
- Each of the first calculation unit 110 and the second calculation unit 120 may be realized, for example, by the above-mentioned processor 11 (see FIG. 1).
- the first calculation unit 110 is configured to be able to calculate an individual likelihood ratio based on two consecutive elements included in the sequence data.
- the individual likelihood ratio is calculated as a likelihood ratio indicating the likelihood of the class to which the two consecutive elements belong.
- the first calculation unit 110 may, for example, sequentially acquire elements included in the sequence data from the data acquisition unit 50, and sequentially calculate individual likelihood ratios based on the two consecutive elements.
- the individual likelihood ratios calculated by the first calculation unit 110 are configured to be output to the second calculation unit 120.
- the second calculation unit 120 is configured to be able to calculate an integrated likelihood ratio based on the multiple individual likelihood ratios calculated by the first calculation unit 110.
- the integrated likelihood ratio is calculated as a likelihood ratio indicating the likelihood of a class to which the multiple elements considered in each of the multiple individual likelihood ratios belong.
- the integrated likelihood ratio is calculated as a likelihood ratio indicating the likelihood of a class to which sequence data consisting of multiple elements belongs.
- the integrated likelihood ratio calculated by the second calculation unit 120 is configured to be output to the class classification unit 200.
- the class classification unit 200 performs class classification of the sequence data based on the integrated likelihood ratio.
- Fig. 11 is a flowchart showing the flow of operations of the information processing device according to the fourth embodiment.
- the data acquisition unit 50 first acquires elements included in the sequence data (step S21).
- the data acquisition unit 50 outputs the acquired elements of the sequence data to the first calculation unit 110.
- the first calculation unit 110 calculates an individual likelihood ratio based on the two consecutive elements obtained (step S22).
- the second calculation unit 120 calculates an integrated likelihood ratio based on the multiple individual likelihood ratios calculated by the first calculation unit 110 (step S23).
- the score calculation unit 150 calculates a score by adding the urgency signal to the integrated likelihood ratio (step S24). Then, the class classification unit 200 performs class classification based on the calculated integrated likelihood ratio (step S25). The class classification may determine one class to which the sequence data belongs, or may determine multiple classes to which the sequence data is likely to belong.
- an individual likelihood ratio is first calculated based on two elements, and then an integrated likelihood ratio is calculated based on multiple individual likelihood ratios.
- an urgency signal is added to calculate the score, so that early class classification can be achieved.
- the information processing device 10 according to the fifth embodiment will be described with reference to Fig. 12 and Fig. 13.
- the fifth embodiment differs from the fourth embodiment in some configurations and operations, and other parts may be the same as the fourth embodiment. Therefore, the following will describe in detail the parts that differ from the embodiments already described, and will omit descriptions of other overlapping parts as appropriate.
- Fig. 12 is a block diagram showing the functional configuration of the information processing device according to the fifth embodiment.
- the same reference numerals are used to designate the same elements as those shown in Fig. 10.
- the information processing device 10 is configured to include a data acquisition unit 50, a likelihood ratio calculation unit 100, a score calculation unit 150, and a class classification unit 200 as components for realizing its functions.
- the likelihood ratio calculation unit 100 includes a first calculation unit 110 and a second calculation unit 120, as in the fourth embodiment described above.
- the first calculation unit 110 includes an individual likelihood ratio calculation unit 111 and a first storage unit.
- the second calculation unit 120 includes an integrated likelihood ratio calculation unit 121 and a second storage unit 122.
- Each of the individual likelihood ratio calculation unit 111 and the integrated likelihood ratio calculation unit 121 may be a processing block realized by, for example, the above-mentioned processor 11 (see FIG. 1).
- Each of the first storage unit 112 and the second storage unit 122 may be realized by, for example, the above-mentioned storage device 14 (see FIG. 1).
- the individual likelihood ratio calculation unit 111 is configured to be able to calculate an individual likelihood ratio based on two consecutive elements among the elements sequentially acquired by the data acquisition unit 50. More specifically, the individual likelihood ratio calculation unit 111 calculates an individual likelihood ratio based on a newly acquired element and past data stored in the first storage unit 112. The information stored in the first storage unit 112 is configured to be readable by the individual likelihood ratio calculation unit 111. When the first storage unit 112 stores past individual likelihood ratios, the individual likelihood ratio calculation unit 111 may read the stored past individual likelihood ratios and calculate new individual likelihood ratios taking into account the acquired elements. On the other hand, when the first storage unit 112 stores the elements themselves acquired in the past, the individual likelihood ratio calculation unit 111 may calculate the past individual likelihood ratios from the stored past elements and calculate the likelihood ratio for the newly acquired element.
- the integrated likelihood ratio calculation unit 121 is configured to be able to calculate an integrated likelihood ratio based on a plurality of individual likelihood ratios.
- the integrated likelihood ratio calculation unit 121 calculates a new integrated likelihood ratio using the individual likelihood ratios calculated by the individual likelihood ratio calculation unit 111 and past integrated likelihood ratios stored in the second storage unit 122.
- the information stored in the second storage unit 122 i.e., past integrated likelihood ratios is configured to be readable by the integrated likelihood ratio calculation unit 121.
- Fig. 13 is a flowchart showing a flow of a likelihood ratio calculation operation in the information processing device according to the fifth embodiment.
- the individual likelihood ratio calculation unit 111 in the first calculation unit 110 reads out past data from the first storage unit 112 (step S31).
- the past data may be, for example, the processing result in the individual likelihood ratio calculation unit 111 of the element acquired immediately before the element currently acquired by the data acquisition unit 50 (in other words, the individual likelihood ratio calculated for the element immediately before).
- the past data may be the element itself acquired immediately before the element acquired in acquisition.
- the individual likelihood ratio calculation unit 111 calculates a new individual likelihood ratio (i.e., an individual likelihood ratio for the element currently acquired by the data acquisition unit 50) based on the elements acquired by the data acquisition unit 50 and the past data read from the first storage unit 112 (step S32).
- the individual likelihood ratio calculation unit 111 outputs the calculated individual likelihood ratio to the second calculation unit 120.
- the individual likelihood ratio calculation unit 111 may store the calculated individual likelihood ratio in the first storage unit 112.
- the integrated likelihood ratio calculation unit 121 in the second calculation unit 120 reads out the past integrated likelihood ratio from the second storage unit 122 (step S33).
- the past integrated likelihood ratio may be, for example, the processing result in the integrated likelihood ratio calculation unit 121 for the element acquired immediately before the element currently acquired by the data acquisition unit 50 (in other words, the integrated likelihood ratio calculated for the previous element).
- the integrated likelihood ratio calculation unit 121 calculates a new integrated likelihood ratio (i.e., an integrated likelihood ratio for the element currently acquired by the data acquisition unit 50) based on the likelihood ratio calculated by the individual likelihood ratio calculation unit 111 and the past integrated likelihood ratio read from the second storage unit 122 (step S34).
- the integrated likelihood ratio calculation unit 121 outputs the calculated integrated likelihood ratio to the class classification unit 200.
- the integrated likelihood ratio calculation unit 121 may store the calculated integrated likelihood ratio in the second storage unit 122.
- individual likelihood ratios are calculated using past individual likelihood ratios, and then an integrated likelihood ratio is calculated using past integrated likelihood ratios.
- the integrated likelihood ratio calculated in this way it is possible to appropriately select the class to which the sequence data belongs.
- the score is calculated by adding an urgency signal as described above, so that early class classification can be achieved.
- each embodiment also includes a processing method in which a program that operates the configuration of each embodiment to realize the functions of the above-mentioned embodiments is recorded on a recording medium, the program recorded on the recording medium is read as code, and executed on a computer.
- computer-readable recording media are also included in the scope of each embodiment.
- each embodiment includes not only the recording medium on which the above-mentioned program is recorded, but also the program itself.
- the recording medium may be, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, magnetic tape, non-volatile memory card, or ROM.
- the scope of each embodiment is not limited to programs recorded on the recording medium that execute processes by themselves, but also includes programs that operate on an OS in conjunction with other software or the functions of an expansion board to execute processes.
- the program itself may be stored on a server, and part or all of the program may be made downloadable from the server to a user terminal.
- the program may be provided to the user in, for example, a SaaS (Software as a Service) format.
- the information processing device described in Supplementary Note 1 includes an acquisition means for acquiring multiple elements included in sequence data, a likelihood ratio calculation means for inputting the multiple elements and calculating a relationship between each element to calculate a likelihood ratio indicating the likelihood of a class to which the sequence data belongs, a score calculation means for adding a value corresponding to an urgency signal to the likelihood ratio to calculate a score, and a classification means for classifying the sequence data into at least one of multiple classes that are classification candidates based on the score.
- the information processing device according to Supplementary Note 2 is the information processing device according to Supplementary Note 1, in which a maximum time allowed from when acquisition of the plurality of elements starts until when the sequence data is classified is determined in advance.
- the information processing device described in Supplementary Note 3 is the information processing device described in Supplementary Note 2, further comprising a determination means for determining whether or not the device is in an emergency state in which it is possible that the sequence data cannot be classified before the maximum time has elapsed, and the score calculation means adds a value corresponding to the urgency signal to the likelihood ratio if the device is in the emergency state, and does not add a value corresponding to the urgency signal to the likelihood ratio if the device is not in the emergency state.
- the information processing device described in Supplementary Note 4 is the information processing device described in any one of Supplementary Notes 1 to 3, further comprising a generation means for generating a value corresponding to the urgency signal using a trained model.
- the information processing device described in Supplementary Note 5 is the information processing device described in Supplementary Note 4, in which the trained model is trained using a loss function plus a value corresponding to a time elapsed since acquisition of the multiple elements began.
- the information processing method described in Supplementary Note 6 is an information processing method which, by at least one computer, acquires a plurality of elements included in sequence data, inputs the plurality of elements and calculates a relationship between the elements to calculate a likelihood ratio indicating the likelihood of a class to which the sequence data belongs, calculates a score by adding a value corresponding to an urgency signal to the likelihood ratio, and classifies the sequence data into at least one of a plurality of classes that are classification candidates based on the score.
- the recording medium described in Supplementary Note 7 is a recording medium having recorded thereon a computer program for causing at least one computer to execute an information processing method of acquiring multiple elements included in sequence data, inputting the multiple elements and calculating relationships between the elements to calculate a likelihood ratio indicating the likelihood of a class to which the sequence data belongs, calculating a score by adding a value corresponding to an urgency signal to the likelihood ratio, and classifying the sequence data into at least one of multiple classes that are classification candidates based on the score.
- the computer program described in Supplementary Note 8 is a computer program that causes at least one computer to execute an information processing method of acquiring multiple elements included in sequence data, inputting the multiple elements and calculating relationships between the elements to calculate a likelihood ratio indicating the likelihood of a class to which the sequence data belongs, calculating a score by adding a value corresponding to an urgency signal to the likelihood ratio, and classifying the sequence data into at least one of multiple classes that are classification candidates based on the score.
- the information processing system described in Supplementary Note 9 is an information processing system including: an acquiring means for acquiring a plurality of elements included in sequence data; a likelihood ratio calculating means for inputting the plurality of elements and calculating a relationship between the elements to calculate a likelihood ratio indicating the likelihood of a class to which the sequence data belongs; a score calculating means for adding a value corresponding to an urgency signal to the likelihood ratio to calculate a score; and a classifying means for classifying the sequence data into at least one of a plurality of classes that are classification candidates based on the score.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2024571478A JP7882351B2 (ja) | 2023-01-17 | 2023-01-17 | 情報処理装置、情報処理方法、及び記録媒体 |
| PCT/JP2023/001187 WO2024154230A1 (ja) | 2023-01-17 | 2023-01-17 | 情報処理装置、情報処理方法、及び記録媒体 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/JP2023/001187 WO2024154230A1 (ja) | 2023-01-17 | 2023-01-17 | 情報処理装置、情報処理方法、及び記録媒体 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2024154230A1 true WO2024154230A1 (ja) | 2024-07-25 |
Family
ID=91955521
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2023/001187 Ceased WO2024154230A1 (ja) | 2023-01-17 | 2023-01-17 | 情報処理装置、情報処理方法、及び記録媒体 |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP7882351B2 (cg-RX-API-DMAC7.html) |
| WO (1) | WO2024154230A1 (cg-RX-API-DMAC7.html) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021229654A1 (ja) * | 2020-05-11 | 2021-11-18 | 日本電気株式会社 | 判定装置、判定方法および記録媒体 |
| WO2022259429A1 (ja) * | 2021-06-09 | 2022-12-15 | 日本電気株式会社 | 情報処理装置、情報処理方法、医療映像識別装置及びプログラムが格納された非一時的なコンピュータ可読媒体 |
-
2023
- 2023-01-17 WO PCT/JP2023/001187 patent/WO2024154230A1/ja not_active Ceased
- 2023-01-17 JP JP2024571478A patent/JP7882351B2/ja active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021229654A1 (ja) * | 2020-05-11 | 2021-11-18 | 日本電気株式会社 | 判定装置、判定方法および記録媒体 |
| WO2022259429A1 (ja) * | 2021-06-09 | 2022-12-15 | 日本電気株式会社 | 情報処理装置、情報処理方法、医療映像識別装置及びプログラムが格納された非一時的なコンピュータ可読媒体 |
Non-Patent Citations (1)
| Title |
|---|
| ACHENCHABE YOUSSEF; BONDU ALEXIS; CORNUéJOLS ANTOINE; DACHRAOUI ASMA: "Early classification of time series", MACHINE LEARNING, SPRINGER US, NEW YORK, vol. 110, no. 6, 1 June 2021 (2021-06-01), New York, pages 1481 - 1504, XP037484648, ISSN: 0885-6125, DOI: 10.1007/s10994-021-05974-z * |
Also Published As
| Publication number | Publication date |
|---|---|
| JP7882351B2 (ja) | 2026-06-30 |
| JPWO2024154230A1 (cg-RX-API-DMAC7.html) | 2024-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210074315A1 (en) | Augmented multi-tier classifier for multi-modal voice activity detection | |
| CN110473525B (zh) | 获取语音训练样本的方法和装置 | |
| CN116204726B (zh) | 一种基于多模态模型的数据处理方法、装置及设备 | |
| JP6176041B2 (ja) | 情報処理装置及びプログラム | |
| JP7740389B2 (ja) | 情報処理装置、情報処理方法、及び記録媒体 | |
| JP7586172B2 (ja) | 情報処理装置およびプログラム | |
| KR102901301B1 (ko) | 전자 장치 및 이의 제어 방법 | |
| EP3834182A1 (en) | Systems and methods to adapt and optimize human-machine interaction using multimodal user-feedback | |
| US11762451B2 (en) | Methods and apparatus to add common sense reasoning to artificial intelligence in the context of human machine interfaces | |
| JP2020529680A (ja) | 通話中の感情を認識し、認識された感情を活用する方法およびシステム | |
| US11087746B2 (en) | Information processing device, information processing method, and program | |
| WO2019127940A1 (zh) | 视频分类模型训练方法、装置、存储介质及电子设备 | |
| CN116964623A (zh) | 数据生成装置、数据生成方法以及程序 | |
| KR20210136724A (ko) | 다중 스케일 예측 손실 함수를 이용한 오디오 신호 학습장치 및 방법 | |
| CN112597912B (zh) | 一种会议内容的记录方法、装置、设备及存储介质 | |
| JP7508333B2 (ja) | 計算機システム及び学習方法 | |
| CN111027358A (zh) | 一种基于书写进度的听写报读方法及电子设备 | |
| US11451694B1 (en) | Mitigation of obstacles while capturing media content | |
| WO2024079853A1 (ja) | 情報処理装置、情報処理方法、及び記録媒体 | |
| WO2025069242A1 (ja) | 情報処理装置、情報処理方法、及び記録媒体 | |
| US20240013407A1 (en) | Information processing apparatus, information processing method, and non-transitory computer-readable storage medium | |
| JP7670161B2 (ja) | 発話推定装置、発話推定方法及び発話推定プログラム | |
| WO2024079854A1 (ja) | 情報処理装置、情報処理方法、及び記録媒体 | |
| WO2025069245A1 (ja) | 情報処理装置、情報処理方法、及び記録媒体 | |
| WO2025069149A1 (ja) | 情報処理装置、情報処理方法、及び記録媒体 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23917444 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 2024571478 Country of ref document: JP Kind code of ref document: A |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2024571478 Country of ref document: JP |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 23917444 Country of ref document: EP Kind code of ref document: A1 |