Abstract: A learning system according to an embodiment includes a model generation device and n calculation devices. The model generation device includes a splitting unit, a secret sharing unit, and a share transmission unit. The splitting unit splits m×n pieces of training data into n groups each including m training data pieces, the n groups corresponding to the respective n calculation devices on one-to-one basis. The secret sharing unit generates m distribution training data pieces for each of the n groups by distributing using a secret sharing scheme and generates distribution training data for each of the m training data pieces in an i-th group among the n groups, using an i-th element Pi among n elements P1, P2, . . . , Pi, . . . , Pn, by distributing using the secret sharing scheme. The share transmission unit transmits corresponding m distribution training data pieces to each of the n calculation devices.

Abstract: An AD conversion circuit includes a plurality of first AD converters, a plurality of voltage holding circuits, a second AD converter, and a selector circuit. Each of the plurality of first AD converters includes a first input terminal. Each of the plurality of first AD converters is a delta-sigma AD converter. Each of the plurality of voltage holding circuits is connected to corresponding one of the plurality of first AD converters. The second AD converter is a delta-sigma AD converter and includes a second input terminal. The selector circuit is connected to the plurality of voltage holding circuits and to the second input terminal. The selector circuit is configured to select, from among the plurality of voltage holding circuits, a voltage holding circuit to be connected to the second input terminal.

Abstract: According to an embodiment, an information processing device is configured to assign a first computing device one or more first tasks of processing respective one or more first partial data of a plurality of partial data included in an n-dimensional target data, n being an integer greater than or equal to 2, the target data being to be processed using a neural network, the one or more first partial data including first data and second data adjacent to the first data in a direction of m-dimension, m being an integer satisfying 1?m?n; and instruct the first computing device to execute a second task included in the one or more first tasks, according to an execution status of second partial data of the plurality of partial data included in the target data, the second partial data being executed by the second computing device.

Abstract: According to one embodiment, a secure computing method includes setting a coefficient selected from a ring of integers Q based on first data X, generating n pieces of first fragment data from the first data X based on the coefficient, causing a learning model held in the computing device to learn the first fragment data, generating n pieces of second fragment data from second data Z based on the coefficient, performing, by each of the n computing devices, inference based on the second fragment data using the learning model, and obtaining decoded data dec by decoding k pieces of inference result data. The coefficient is set to make each of the n pieces of first fragment data less than a maximum value of the ring of integers Q.

Abstract: According to one embodiment, each of the client terminals includes a first processor configured to execute a learning process of machine learning model, extract a first parameter column from the machine learning model, change the arrangement of parameters, perform secret sharing with respect to the first parameter column, and transmit a first fragment parameter column. Each of the aggregated server device includes a second processor configured to receive first fragment parameter columns, change arrangement of fragment parameters, and execute an aggregation process. The machine learning model is updated based on parameters in a second parameter column decoded from second fragment parameter columns generated in the aggregated server devices.

Abstract: An information processing system includes a server device and one or more edge systems. The server device includes a memory, and one or more processors coupled to the memory and configured to: receive a plurality of first stream data to which first time information is associated from the edge systems; receive reference information in which second time information and the first time information corresponding to the selected first stream data are associated to each other, from the edge systems; receive an acquisition request of the first stream data corresponding to third time information expressed in the same format as the second time information; and identify the first stream data to which the first time information corresponding to the third time information is associated based on the reference information.

Abstract: An information processing system includes a server device and one or more edge systems. The server device includes a memory, and one or more processors coupled to the memory and configured to: receive a plurality of first stream data to which first time information is associated from the edge systems; receive reference information in which second time information and the first time information corresponding to the selected first stream data are associated to each other, from the edge systems; receive an acquisition request of the first stream data corresponding to third time information expressed in the same format as the second time information; and identify the first stream data to which the first time information corresponding to the third time information is associated based on the reference information.

Abstract: According to one embodiment, a secure computing method includes setting a coefficient selected from a ring of integers Q based on first data X, generating n pieces of first fragment data from the first data X based on the coefficient, causing a learning model held in the computing device to learn the first fragment data, generating n pieces of second fragment data from second data Z based on the coefficient, performing, by each of the n computing devices, inference based on the second fragment data using the learning model, and obtaining decoded data dec by decoding k pieces of inference result data. The coefficient is set to make each of the n pieces of first fragment data less than a maximum value of the ring of integers Q.

Abstract: According to an embodiment, an information processing device is configured to assign a first computing device one or more first tasks of processing respective one or more first partial data of a plurality of partial data included in an n-dimensional target data, n being an integer greater than or equal to 2, the target data being to be processed using a neural network, the one or more first partial data including first data and second data adjacent to the first data in a direction of m-dimension, m being an integer satisfying 1?m?n; and instruct the first computing device to execute a second task included in the one or more first tasks, according to an execution status of second partial data of the plurality of partial data included in the target data, the second partial data being executed by the second computing device.

Abstract: A learning system according to an embodiment includes a model generation device and n calculation devices. The model generation device includes a splitting unit, a secret sharing unit, and a share transmission unit. The splitting unit splits m×n pieces of training data into n groups each including m training data pieces, the n groups corresponding to the respective n calculation devices on one-to-one basis. The secret sharing unit generates m distribution training data pieces for each of the n groups by distributing using a secret sharing scheme and generates distribution training data for each of the m training data pieces in an i-th group among the n groups, using an i-th element Pi among n elements P1, P2, . . . , Pi, . . . , Pn, by distributing using the secret sharing scheme. The share transmission unit transmits corresponding m distribution training data pieces to each of the n calculation devices.

Abstract: According to one embodiment, an anomaly detection device includes predicted value calculation unit, an anomaly degree calculation unit, a second predicted value calculation unit, a determination value calculation unit, and an anomaly determination unit. The first predicted value calculation unit calculates a first model predicted value from a correlation model obtained by first machine learning, the anomaly degree calculation unit calculates an anomaly degree, the second predicted value calculation unit calculates a second model predicted value from a time series model obtained by second machine learning, the determination value calculation unit calculates a divergence degree, and the anomaly determination unit determines whether an anomaly occurs or not.

Abstract: An information processing device has a digital-to-pulse converter which outputs a pulse signal including a pulse having a pulse length in accordance with a digital input signal, and a selective oscillator which performs an oscillation operation while the pulse of the pulse signal is output and holds an oscillation operation state at a point of time where the output of the pulse is stopped.

Abstract: According to an embodiment, a radiation measuring apparatus includes a detector, comparators, a threshold controller, counters, and a generator. The detector includes plural detecting elements each configured to convert energy of incident radiation into a first electrical signal. The comparators correspond to the respective detecting elements, each comparator being configured to output a second electrical signal when a level of the corresponding first electrical signal is not less than a threshold. The threshold controller is configured to supply a first value as the threshold to the respective comparators at a first time, and supply a second value as the threshold to the respective comparators at a second time. The counters correspond to the respective comparators, each counter being configured to count the corresponding second electrical signal. The generator is configured to generate a pulse height frequency distribution of the radiation by using counts of the counters.

Abstract: A time to digital converter has a counter to measure the number of cycles of a first signal, a first phase difference detector to generate a phase difference signal having a pulse width corresponding to a phase difference, a first capacitor to be charged with an electric charge, a second capacitor including capacitance N times the capacitance of the first capacitor, the N being a real number larger than 1, a comparator to compare a charge voltage of the first capacitor and a charge voltage of the second capacitor, a first charge controller to continue to charge the second capacitor until the comparator detects that the charge voltage of the second capacitor has reached the charge voltage of the first capacitor or more, and a first phase difference arithmetic unit to operate the phase difference between the first signal and the second signal.

Abstract: A receiver has a receiving unit to receive a radio signal, a signal detector to detect a reception signal in each of a plurality of set periods shifted in time to be overlapped in a partial period, and a demodulating unit to perform demodulation processing based on the reception signal. The signal detector has a smoothing unit to smooth the output signal of the receiving unit in each of the plurality of set periods, a comparing unit to output a signal obtained by comparing a level of the smoothed signal, with a threshold value, and an initializing unit to initialize the signal smoothed by the smoothing processing unit, every time the comparing unit compares the smoothed signal with the threshold value, and the demodulating unit performs the demodulation processing based on the smoothed signal determined to be the threshold value or more by the comparing unit.

Abstract: A waveform shaping filter according to an embodiment includes at least one filter stage and a control circuit. The filter stage includes a differentiation signal generation circuit, a proportional signal generation circuit, and an adder circuit. The differentiation signal generation circuit generates a differentiation signal obtained by amplifying a differentiation component of an input signal. The proportional signal generation circuit generates a proportional signal obtained by amplifying the input signal. The adder circuit outputs an output signal obtained by adding the proportional signal and the differentiation signal. The control circuit compares the output signal and a first detection level, detects at least one of an overshoot and an undershoot of the output signal, and controls a time constant of the filter stage, based on a detection result.

Abstract: An information processing device has a digital-to-pulse converter which outputs a pulse signal including a pulse having a pulse length in accordance with a digital input signal, and a selective oscillator which performs an oscillation operation while the pulse of the pulse signal is output and holds an oscillation operation state at a point of time where the output of the pulse is stopped.

Abstract: A time to digital converter has a counter, a first phase difference detector, a first capacitor, a second capacitor having capacitance N times a capacitance of the first capacitor, a comparator to compare a charge voltage of the first capacitor with a charge voltage of the second capacitor, a first charge controller, a first phase difference arithmetic unit, a second phase difference detector, a second charge controller, a second phase difference arithmetic unit to operate the phase difference between the first signal and the second signal, and a third phase difference arithmetic unit to detect a fractional phase difference between the first signal and the second signal. The first phase difference arithmetic unit operates the phase difference between the first signal and the second signal, based on a reference phase, when the counter suspends a measurement operation.

Abstract: A wireless communication apparatus has a transmitter, a signal processor, and ADPLL circuitry. The transmitter to modulate transmission data using a local oscillation signal to generate a wireless signal to be transmitted from an antenna. The signal processor to generate the transmission data and to supply the generated transmission data to the transmitter. The ADPLL (All Digital Phase-Locked Loop) circuitry to generate the local oscillation signal by ADPLL processing and to supply digital information correlated with an input sensing signal to the signal processor.

Abstract: An oscillator has an oscillator, an integer phase detector, a random number generator, an edge selector, a fractional phase detector, an offset correction arithmetic unit, and a phase error generator. The oscillator generates an oscillation signal having an oscillation frequency in accordance with a phase error signal. The integer phase detector detects an integer phase of the oscillation signal. The random number generator generates a random number. The edge selector outputs a phase difference signal indicating a phase difference between a phase of a reference signal and a phase of the oscillation signal, or indicating a phase difference acquired by offsetting the phase difference. The fractional phase detector detects a fractional phase of the oscillation signal based on the phase difference signal. The offset correction arithmetic unit computes an offset correction value in accordance with the random number. The phase error generator generates the phase error signal.