Kabushiki Kaisha Toshiba

Tomohiro Nakai

Satoshi Ito

Susumu Kubota

Kazunori Imoto

Yasunobu Yamauchi

Tsukasa Ike

Toshimitsu Kaneko

Tomoki Watanabe

Tatsuo Kozakaya

Kanako Nakayama

Sawa Fuke

Yasuyuki Tsunoi

According to one embodiment, a learning apparatus includes a processor. The processor acquires first training data. The processor inputs the first training data to a model, and generate a plurality of estimation vectors that are a processing result of the model. The processor generates an estimation distribution from the estimation vectors. The processor calculates a distribution loss between the estimation distribution and a target distribution that is a target in an inference using the model. The processor updates parameters of the model, based on the distribution loss.

Learning apparatus, method and computer readable medium

According to one embodiment, a defect management apparatus includes a processor. The processor acquires first information and second information, the first information including first defect positions relating to defects detected with a first device for an inspection target and corresponding first labels indicating classifications of the defects, the second information including second defect positions relating to defects detected with a second device for the inspection target. The processor determines a first defect position corresponding to a second defect position as a corresponding defect position. The processor diverts the first label corresponding to the corresponding defect position as a second label of the second defect position.

Defect management apparatus, method and non-transitory computer readable medium

According to one embodiment, an information processing device includes a hardware processor configured to acquire operation cost information indicative of a relationship between a state of an operator and a period of time required for the operator to perform an operation from a storage that stores the operation cost information, acquire state information indicative of a state of a target operator, and calculate a period of time required for the target operator to perform a target operation based on the operation cost information and the state information.

Information processing device and method

A recognition apparatus according to an embodiment of the present disclosure includes a memory and a hardware processor coupled to the memory. The hardware processor is configured to: extract first feature quantity data from sensor data; generate attention information based on a classification contribution of the first feature quantity data; generate a second feature quantity data by processing the first feature quantity data with the attention information; generate processed feature quantity data including the first feature quantity data and the second feature quantity data; and perform classification of a recognition object from the processed feature quantity data by using a classification network.

Recognition apparatus, recognition method, and program product

According to an embodiment, an apparatus includes a calculator, an pixel evaluator, an accumulator, and an area evaluator. The calculator is configured to calculate a feature of an image for each pixel in image data. The pixel evaluator is configured to produce a score that evaluates the feature for each pixel. The accumulator is configured to calculate, for each pixel, a cumulative score obtained by accumulating all scores in an area including a minor angle formed by a half line in a first direction from the each pixel position and another half line in a second direction from the each pixel position. The area evaluator is configured to calculate an evaluation value that is a total of the scores in a quadrilateral area enclosed by two lines of the first direction and two lines of the second direction based on the cumulative scores at pixel positions at vertexes of the quadrilateral area.

Image evaluation apparatus, image evaluation method, and computer program product

According to an embodiment, a computing apparatus includes a memory, and a processor. The memory stores N first vectors in a d-dimensional binary vector space consisting of binary values. The processor acquires a second vector in the d-dimensional binary vector space. The processor extracts M first vectors having a distance from the second vector satisfying a first condition out of the N first vectors, and calculate a distribution of distances of the M first vectors from the second vector. The processor acquires a first kernel function per a first distance between the M first vectors and the second vector in a first range. The processor generates a second kernel function based on the distribution and the first kernel functions. The processor calculates an occurrence probability of the second vector in the N first vectors based on the second kernel function.

Apparatus, method, and computer program product for computing occurrence probability of vector

A recognition device includes a storage unit, an acquiring unit, a first calculator, a second calculator, a determining unit, and an output unit. The storage unit stores multiple training patterns each belonging to any one of multiple categories. The acquiring unit acquires a recognition target pattern to be recognized. The first calculator calculates, for each of the categories, a distance histogram representing distribution of the number of training patterns belonging to the category with respect to distances between the recognition target pattern and the training patterns belonging to the category. The second calculator analyzes the distance histogram of each of the categories to calculate confidence of the category. The determining unit determines a category of the recognition target pattern from the multiple categories by using the confidences. The output unit outputs the category of the recognition target pattern.

Recognition device, method, and computer program product

According to an embodiment, a recognition device includes a generation unit to select, plural times, groups each including learning samples from a storage unit, learn a classification metric for classifying the groups selected in each selection, and generate an evaluation metric including the classification metrics; a transformation unit to transform a first feature value of an image including an object into a second feature value using the evaluation metric; a calculation unit to calculate similarities of the object to categories in a table using the second feature value and reference feature values; and a registration unit to register the second feature value as the reference feature value in the table associated with the category of the object and register the first feature value as the learning sample belonging to the category of the object in the storage unit. The generation unit performs the generation again.

Growing community of inventors

Kanagawa, Japan

Tomohiro Nakai