Kioxia Corporation

Toshiba Memory Corporation

Youyang Ng

Koji Horisaki

Ryo Yamaki

Gibeom Park

Kazuhisa Horiuchi

Takeshi Fujiwara

Takuji Ohashi

Kenichi Maeda

Masahiro Ishiyama

Yusuke Doi

Takaomi Murakami

Yosuke Bando

Osamu Yamane

Yujen Lai

Kengo Nakata

Chihiro Ida

Takeshi Sakoda

Yuko Kono

Atsushi Nakajima

Bo Wang

Yuchieh Lin

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.

Image processing apparatus and image processing method

An information processing apparatus has an acquisitor configured to acquire an entire area image obtained by capturing an entire area of a processing surface of a wafer including at least one defect, a training image selector configured to select, as a training image, a partial image including at least one defect from the entire area image, a model constructor configured to construct a calculation model of generating a label image obtained by extracting and binarizing the defect included in the partial image, and a learner configured to update a parameter of the calculation model based on a difference between the label image generated by inputting the training image to the calculation model and a reference label image obtained by extracting and binarizing the defect of the training image.

Information processing apparatus and information processing method

A semiconductor image processing apparatus has an image input unit inputs a first semiconductor image, an exposure condition input unit configured to input exposure conditions, a generator performs a process of extracting a feature amount in consideration of the exposure conditions while reducing resolution of the first semiconductor image and thereafter use the extracted feature amount to increase the resolution to generate a second semiconductor image, and a discriminator configured to discriminate whether the input image is the second semiconductor image or a third semiconductor image provided in advance. The generator performs learning so that the discriminator erroneously discriminates the second semiconductor image as the third semiconductor image based on a result discriminated by the discriminator. The discriminator performs learning so as not to erroneously discriminate the second semiconductor image as the third semiconductor image, and not to erroneously discriminate the third semiconductor image as the second semiconductor image.

Semiconductor image processing apparatus

A memory system includes a non-volatile memory having a plurality of memory cells and a memory controller. The memory controller is configured to generate a histogram indicating, with respect to each of a plurality of threshold voltage levels for multi-level cell (MLC) reading, a number of memory cells at the threshold voltage level, based on data read from the plurality of memory cells using a plurality of reference read voltages, estimate a plurality of read voltages for MLC reading of the plurality of memory cells as estimation values by inputting the histogram into a read voltage estimation model, determine, through MLC reading of the plurality of memory cells using a plurality of sets of read voltages, a set of read voltages for MLC reading as observation values, and update one or more parameters of the read voltage estimation model based on the estimation values and the observation values.

Memory system

According to one embodiment, in a memory system, a memory controller is configured to execute a first operation of observing an optimum value of a read voltage and updating a set value based on the observation result of the optimum value, at a predetermined time point of a plurality of time points for updating the set value of the read voltage for a plurality of memory cells, and execute a second operation of updating the set value based on the set value updated at one previous time point without executing the observation of the optimum value, at a time point after one time point of the predetermined time point.

According to an embodiment, a memory system includes a plurality of memory cells and a memory controller. The memory controller executes a read operation on the plurality of memory cells using one or more reference read voltages to acquire a histogram representing the number of memory cells with respect to a threshold voltage. The memory controller inputs the histogram to a trained neural network model including an output layer for outputting one or more actual read voltages to read data from the memory cells. The memory controller executes a read operation on the memory cells using the one or more actual read voltages output from the output layer.

Memory system and method

A memory controller performs a reference read on a plurality of memory cells using reference read voltages, generates a histogram indicating the number of memory cells in different threshold voltage bins based on results of the reference read, estimates actual read voltages based on the histogram and a first estimation function, and reads data using the actual read voltages. When reading of the data with the actual read voltages estimated using the first estimation function fails, the memory controller estimates actual read voltages using a second estimation function different from the first estimation function and reads the data with the actual read voltages estimated using the second estimation function.

According to one embodiment, a communication apparatus includes a congestion degree calculating unit and an operation mode setting unit. The congestion degree calculating unit calculates a third congestion degree related to wireless communication based on a first congestion degree of first wireless communication and a second congestion degree of second wireless communication. In the first wireless communication, peer-to-peer communication is performed directly between communication apparatuses, and in the second wireless communication, communication is performed through a wireless communication device. The operation mode setting unit switches an operation mode of the first wireless communication based on the third congestion degree to either of a mode in which the communication apparatus transmits content or a mode in which the communication apparatus receives content.

Growing community of inventors

Yokohama, Japan

Youyang Ng