Abstract: An arithmetic operation device is provided that removes a part of parameters of a predetermined number of parameters from a first machine learning model which includes the predetermined number of parameters and is trained so as to output second data corresponding to input first data, determines the number of bits of a weight parameter according to required performance related to an inference to generate a second machine learning model, and acquires data output from the second machine learning model so as to correspond to the input first data with a smaller computational complexity than the first machine learning model.

Abstract: An image processing apparatus has a first image acquisitor that acquires a source image, a second image acquisitor that acquires a first target image, a label acquisitor that acquires a label, a feature extractor including a first neural network that extracts a feature of the source image and a feature of the first target image, a class classifier including a second neural network that performs a class classification of the source image and the first target image, a domain classifier including a third neural network that performs a domain classification of the source image and the first target image, a processor that assigns a pseudo label to the first target image, a self-learner that performs a self-learning of the first neural network, the second neural network, and the third neural network, and a learner that learns the first, second and third neural networks, by performing a back propagation process.

Abstract: According to one embodiment, an information processing method includes: calculating a first feature amount of query data in a first field; calculating first similarity degrees between the first feature amount and second feature amounts in the first field; obtaining, based on the first similarity degrees, third feature amounts in a second field that are associated with feature amounts selected from the second feature amounts, the second field being different from the first field; calculating fourth feature amounts in the second field, for choices concerning the query data; calculating second similarity degrees between the third feature amounts and the fourth feature amounts; and selecting, based on the second similarity degrees, an answer to the query data among answer candidates corresponding to the third feature amounts.

Abstract: According to one embodiment, a processor is configured to calculate a calculation amount in inference time of a neural network, using a result of summing, with respect to a group to which quantization is applied, products of the number of product-sum operations and bit widths of weight for the product-sum operations in the neural network. Then, the processor is configured to optimize a value of the weight and a quantization step size to minimize the recognition error by the neural network based on the calculated calculation amount, and execute computing about the neural network based on the optimized weight and the quantization step size.

Abstract: According to one embodiment, an arithmetic operation device removes a part of parameters of a predetermined number of parameters from a first model which includes the predetermined number of parameters and is trained so as to output second data corresponding to input first data and determines the number of bits of a weight parameters according to required performance related to an inference to generate a second model, and acquires data output from the second model so as to correspond to the input first data with a smaller computational complexity than the first model.

Abstract: According to one embodiment, a training device includes a first memory, a second memory, and a processing circuit. The first memory is a memory accessible at a higher speed than the second memory. The training device executes a training process of a machine learning model using a stochastic gradient descent method. The processing circuit stores a first output produced by the process of a first layer in the second memory, and stores a second output produced by the process of a second layer, in a forward process of the training process. The processing circuit updates a parameter of the second layer based on the second output stored in the first memory, reads the first output stored in the second memory, and updates a parameter of the first layer based on the read first output, in a backward process of the training process.

Abstract: According to one embodiment, an information processing apparatus for convolution operations in layers of a convolutional neural network, includes a memory and a product-sum operating circuitry. The memory is configured to store items of information indicative of an input, a weight to the input, and a bit width determined for each filter of the weight. The product-sum operating circuitry is configured to perform a product-sum operation based on the items of information indicative of the input, the weight, and the bit width, stored in the memory.

Abstract: According to one embodiment, an arithmetic device, includes a first processing layer and a second processing layer, each configured to perform an arithmetic operation on input data and constituting a part of a multi-layer neural network configured to perform corrections by an error backward propagation scheme; a detour path that connects an input and an output of the second processing layer; and an evaluation unit configured to evaluate operation results of the first and the second processing layers.

Abstract: An intercom system for a building or apartment house includes at least one main entrance station 1 which is installed at an entrance or the like of a building or an apartment house, a plurality of residential master stations 2 which are respectively installed inside a plurality of residences in the building or apartment house for receiving calls and talking with the main entrance station 1, a common line 3 which connects the main entrance station 1 and respective residential master stations 2 in parallel, a line power supply 4 which supplies direct current power, for example, DC 24 V to the main entrance station 1 and respective residential master stations 2 through the common line 3, an electronic door lock 5 which is connected to the main entrance station 1 for locking and releasing an entrance door of a main entrance, and a plurality of call switches 6 connected to respective residential master stations 2 and mounted on front doors of respective residences for operation by a visitor.

Abstract: An intercom system for mansion/apartment house of the present invention can achieve the power saving by suppressing a current consumption at the time of calling by a visitor standing at a main entrance or a residential front door or at the time of locking or releasing an electric door lock which is mounted on a front door of the main entrance and can ensure the favorable talk quality by suppressing howling. To a common bus, at least one main entrance station including a talk module for performing calling and talking and at least one switch module being connected to the talk module and having a call switch for performing selecting of a station, a bus controller for supplying a power to a common bus, and a plurality of residential master stations for performing calling and talking with the main entrance station are connected in parallel respectively.