Abstract: An anomaly detection method by which a computer performs anomaly detection includes: obtaining first feature data outputted through N (N is an integer not less than 1) convolutional layers of a convolutional neural network configured as an encoder when an image is inputted to the convolutional neural network; obtaining second feature data outputted through M (M is an integer not less than 1, and M?N) convolutional layers of the convolutional neural network and different in size from the first feature data; and performing anomaly detection on the image by using features indicated by the first feature data and the second feature data that are different in size.

Abstract: A training method performed through batch learning by a computer includes: obtaining training data including first time-series data and second time-series data different from the first time-series data; performing first training processing of training a neural process (NP) model, which outputs, using a stochastic process, a prediction result that takes uncertainty into account, to predict first and second time-series data distributions, based on the first time-series data and second time-series data; and performing, using a contrastive learning algorithm, second training processing of (i) training the NP model to bring close to each other first sampling data items generated by sampling from the first time-series data distribution, (ii) training the NP model to bring close to each other second sampling data items generated by sampling from the second time-series data distribution, and (iii) training the NP model to push away the first and second sampling data items far from each other.

Abstract: First optimization processing for optimizing parameters of a DNN and second optimization processing for optimizing hyperpararneters for each sample used in data augmentation processing are alternately performed. The first optimization processing includes causing the DNN to predict a first augmentation label from a first augmented sample, calculating a first error function between the first augmentation label and a first correct label for a first sample, and updating the parameters in accordance with the first error function. The second optimization processing includes acquiring a second sample, causing the DNN after the updating of the parameters to predict a second label from the second sample, calculating a second error function between the second label and a second correct label for the second sample, and updating the hyperparameter in accordance with a gradient obtained by differentiation of the second error function with respect to the hyperparameter.

Abstract: Executing a deep neural network by obtaining, during deep neural network inference, a binary intermediate feature map in binary representation by converting a floating-point or fixed-point intermediate feature map into a binary vector using a first transformation module; generating a compressed feature map by compressing the binary intermediate feature map using a nonlinear dimensionality reduction layer; storing the compressed feature map into memory; reconstructing the binary intermediate feature map by decompressing the compressed feature map read from the memory using a reconstruction layer corresponding to the nonlinear dimensionality reduction layer; and converting the reconstructed binary intermediate feature map into a floating-point or fixed-point intermediate feature map using a second transformation module.

Abstract: An information processing method performed by a computer includes: obtaining a plurality of recognition result candidates in sensing data and a likelihood of each of the plurality of recognition result candidates, the plurality of recognition result candidates and the likelihood being obtained by inputting the sensing data to a model that is trained by machine learning and performs recognition processing; obtaining an indication designating a part to be analyzed in the sensing data; selecting at least one recognition result candidate from the plurality of recognition result candidates, based on (i) a relationship between each of the plurality of recognition result candidates and the part and (ii) the likelihood of each of the plurality of recognition result, candidates; and outputting the at least one recognition result candidate that is selected.

Abstract: What is disclosed is a method for wireless communication comprising receiving a wireless communication via a receiver of the mobile communication device, deriving a demodulation reference signal from a first plurality of symbols of the wireless communication; creating a channel estimation matrix using the demodulation reference signal; inverting the channel estimation matrix to obtain a channel pseudo-inverse matrix; deriving a tracking reference signal from a second plurality of symbols of the wireless communication; calculating a phase shift for one or more additional symbols based on the tracking reference signal; determining a corrected channel pseudo-inverse matrix for the one or more additional symbols by adjusting the channel pseudo-inverse matrix according to the calculated phase shift; and controlling the receiver to accomplish data detection using the corrected channel pseudo-inverse matrix on one or more orthogonal frequency division multiplexing subcarriers.

Abstract: An information processing method implemented by a computer includes: obtaining a piece of first data, and a piece of second data not included in a training dataset for training an inferencer; calculating, using a piece of first relevant data obtained by inputting the first data to the inferencer trained by machine learning using the training dataset, a first contribution representing contributions of portions constituting the first data to a piece of first output data output by inputting the first data to the inferencer; calculating, using a piece of second relevant data obtained by inputting the second data to the inferencer, a second contribution representing contributions of portions constituting the second data to a piece of second output data output by inputting the second data to the inferencer; and determining whether to add the second data to the training dataset, according to the similarity between the first and second contributions.

Abstract: An information processing method includes: inputting an input tensor indicating data to a processor having a memory; causing the processor to perform, after elements of the input tensor are subjected to precomputation for conversion into a power-of-two format and are stored in the memory, convolution operation processing with only addition and shift operations by using the precomputed elements of the input tensor stored in the memory and weight tensors that are pre-converted into the power-of-two format in accordance with a predetermined algorithm, that are stored in the memory, and that indicate weights having a possibility of being used for a convolution operation; and outputting, as an output tensor, the elements of the input tensor on which the convolution operation processing is performed.

Abstract: An information processing method performed by a computer includes: obtaining a plurality of recognition result candidates in sensing data and a likelihood of each of the plurality of recognition result candidates, the plurality of recognition result candidates and the likelihood being obtained by inputting the sensing data to a model that is trained by machine learning and performs recognition processing; obtaining an indication designating a part to be analyzed in the sensing data; selecting at least one recognition result candidate from the plurality of recognition result candidates, based on (i) a relationship between each of the plurality of recognition result candidates and the part and (ii) the likelihood of each of the plurality of recognition result, candidates; and outputting the at least one recognition result candidate that is selected.

Abstract: Executing a deep neural network by obtaining, during deep neural network inference, a binary intermediate feature map in binary representation by converting a floating-point or fixed-point intermediate feature map into a binary vector using a first transformation module; generating a compressed feature map by compressing the binary intermediate feature map using a nonlinear dimensionality reduction layer; storing the compressed feature map into memory; reconstructing the binary intermediate feature map by decompressing the compressed feature map read from the memory using a reconstruction layer corresponding to the nonlinear dimensionality reduction layer; and converting the reconstructed binary intermediate feature map into a floating-point or fixed-point intermediate feature map using a second transformation module.

Abstract: An improved sliding discrete Fourier transform (SDFT) algorithm called CIC-SDFT and its apparatus are provided. An input signal is multiplied by a modulated twiddle factor, and is then processed by a modified cascade integrator-comb (CIC) filter. The CIC-SDFT comprises an integrator section and a comb section, with a downsampler disposed between the two sections to enable downsampling of the SDFT computations. Through addition of more integrator stages and comb stages to the CIC-SDFT, the accuracy of spectrum estimation may be improved in a computationally inexpensive manner and with less complexity than applying windowing functions to known SDFTs. Various embodiments provide a partially-nonrecursive method of CIC-SDFT that further decreases computational complexity.

Abstract: What is disclosed is a method for wireless communication comprising receiving a wireless communication via a receiver of the mobile communication device, deriving a demodulation reference signal from a first plurality of symbols of the wireless communication; creating a channel estimation matrix using the demodulation reference signal; inverting the channel estimation matrix to obtain a channel pseudo-inverse matrix; deriving a tracking reference signal from a second plurality of symbols of the wireless communication; calculating a phase shift for one or more additional symbols based on the tracking reference signal; determining a corrected channel pseudo-inverse matrix for the one or more additional symbols by adjusting the channel pseudo-inverse matrix according to the calculated phase shift; and controlling the receiver to accomplish data detection using the corrected channel pseudo-inverse matrix on one or more orthogonal frequency division multiplexing subcarriers.

Abstract: Systems and methods are provided in which a wireless receiver can be configured to digitally synchronize a receive sampling rate to a transmit sampling rate, and may include a digital interpolator controlled by a timing control unit with a timing offset estimator. The timing control unit can be configured to calculate and output parameters to the digital interpolator. The digital interpolator can include a sample buffer followed by a fractional delay filter. Output parameters to the digital interpolator can include a fractional delay timing offset signal of the receiver relative to a transmitter timing signal and a buffer pointer control signal to control a position of the read pointer relative to a write pointer to compensate for subsample timing offset. The timing offset estimator can be configured to calculate and provide to the timing control unit a sampling period ratio control word and an instantaneous timing offset control word.

Abstract: An information processing method includes: inputting an input tensor indicating data to a processor having a memory; causing the processor to perform, after elements of the input tensor are subjected to precomputation for conversion into a power-of-two format and are stored in the memory, convolution operation processing with only addition and shift operations by using the precomputed elements of the input tensor stored in the memory and weight tensors that are pre-converted into the power-of-two format in accordance with a predetermined algorithm, that are stored in the memory, and that indicate weights having a possibility of being used for a convolution operation; and outputting, as an output tensor, the elements of the input tensor on which the convolution operation processing is performed.

Abstract: What is disclosed is a method for wireless communication comprising receiving a wireless communication via a receiver of the mobile communication device, deriving a demodulation reference signal from a first plurality of symbols of the wireless communication; creating a channel estimation matrix using the demodulation reference signal; inverting the channel estimation matrix to obtain a channel pseudo-inverse matrix; deriving a tracking reference signal from a second plurality of symbols of the wireless communication; calculating a phase shift for one or more additional symbols based on the tracking reference signal; determining a corrected channel pseudo-inverse matrix for the one or more additional symbols by adjusting the channel pseudo-inverse matrix according to the calculated phase shift; and controlling the receiver to accomplish data detection using the corrected channel pseudo-inverse matrix on one or more orthogonal frequency division multiplexing subcarriers.

Abstract: An improved sliding discrete Fourier transform (SDFT) algorithm called CIC-SDFT and its apparatus are provided. An input signal is multiplied by a modulated twiddle factor, and is then processed by a modified cascade integrator-comb (CIC) filter. The CIC-SDFT comprises an integrator section and a comb section, with a downsampler disposed between the two sections to enable downsampling of the SDFT computations. Through addition of more integrator stages and comb stages to the CIC-SDFT, the accuracy of spectrum estimation may be improved in a computationally inexpensive manner and with less complexity than applying windowing functions to known SDFTs. Various embodiments provide a partially-nonrecursive method of CIC-SDFT that further decreases computational complexity.

Abstract: Systems and methods are provided in which a wireless receiver can be configured to digitally synchronize a receive sampling rate to a transmit sampling rate, and may include a digital interpolator controlled by a timing control unit with a timing offset estimator. The timing control unit can be configured to calculate and output parameters to the digital interpolator. The digital interpolator can include a sample buffer followed by a fractional delay filter. Output parameters to the digital interpolator can include a fractional delay timing offset signal of the receiver relative to a transmitter timing signal and a buffer pointer control signal to control a position of the read pointer relative to a write pointer to compensate for subsample timing offset. The timing offset estimator can be configured to calculate and provide to the timing control unit a sampling period ratio control word and an instantaneous timing offset control word.

Abstract: What is disclosed is a method for wireless communication comprising receiving a wireless communication via a receiver of the mobile communication device, deriving a demodulation reference signal from a first plurality of symbols of the wireless communication; creating a channel estimation matrix using the demodulation reference signal; inverting the channel estimation matrix to obtain a channel pseudo-inverse matrix; deriving a tracking reference signal from a second plurality of symbols of the wireless communication; calculating a phase shift for one or more additional symbols based on the tracking reference signal; determining a corrected channel pseudo-inverse matrix for the one or more additional symbols by adjusting the channel pseudo-inverse matrix according to the calculated phase shift; and controlling the receiver to accomplish data detection using the corrected channel pseudo-inverse matrix on one or more orthogonal frequency division multiplexing subcarriers.

Abstract: Systems and methods are provided in which a wireless receiver can be configured to compensate frequency and phase offsets with joint symbol timing recovery of modulated signals transmitted across a channel, and it can include a transformation module configured to generate an error signal for an information signal representing the modulated signal received by the receiver. The transformation module can include a squaring module configured to square the information signal, thereby generating a squared signal, and a mixer configured to perform a complex multiplication of the squared signal by the local reference signal, and a downsampler. The transformation module can also be configured to extract and compensate frequency and phase offsets with joint symbol timing recovery.

Abstract: Systems and methods are provided in which a wireless receiver can be configured to compensate frequency and phase offsets with joint symbol timing recovery of modulated signals transmitted across a channel, and it can include a transformation module configured to generate an error signal for an information signal representing the modulated signal received by the receiver. The transformation module can include a squaring module configured to square the information signal, thereby generating a squared signal, and a mixer configured to perform a complex multiplication of the squared signal by the local reference signal, and a downsampler. The transformation module can also be configured to extract and compensate frequency and phase offsets with joint symbol timing recovery.