Mitsubishi Heavy Industries Limited

International Business Machines Corporation

Mitsubishi Electric Corporation

Fuji Electric Co., Ltd.

Kabushiki Kaisha Square Enix

Other

Aisin Aw Co., Ltd.

Mitsubishi Jukogyo Kabushiki Kaisha

Masaki Ono

Katsunori Tanaka

Kohichi Kamijoh

Kazufumi Ikeda

Shigemi Mandai

Masaharu Nishimura

Keizo Ohnishi

Yasuoki Tomita

Kiyoshi Suenaga

Akira Koseki

Takayuki Katsuki

Mikio Suzuki

Kenta Taniguchi

Keisuke Matsuyama

Tetsu Konishi

Yoshihiko Kobayashi

Shunichi Amano

Masatoshi Kodama

Kouji Watanabe

Nobuhiro Tsukioka

Keisuke Nakamura

Takayuki Osogami

Michiaki Tatsubori

Kenji Suzuki

Shin Soyano

Daisuke Takuma

Tatsuo Ishiguro

Kuniaki Aoyama

Tetsuya Nasukawa

Akiko Fujiuchi

Sosuke Nakamura

Toshiro Takase

Kazuo Aoki

Yohei Ikawa

Akira Morozumi

Taiki Kinoshita

Eiji Akita

Koichi Akagi

Kentaro Tokuyama

Ryo Kawahara

Daiki Kimura

Subhajit Chaudhury

Mitsuru Kondo

Masao Terazaki

Yutaka Kawata

Junji Tsuruoka

Jun Kubota

Naoyuki Tokida

Fumitoshi Kiyooka

Yoshihiro Katsu

Masayuki Takahama

Akihiro Kosugi

Toshinobu Ohara

Hideki Murata

Yasuhiko Ikegami

Sunao Umemura

Mikio Uesugi

Eigo Kato

Masamitsu Okuzono

Takahiro Ishihara

Katsuhiko Yoshihara

Tetsuya Iwamoto

Takaharu Tsugane

Masanori Mori

A computer-implemented method for reinforcement learning with Logical Neural Networks (LNNs) is provided including receiving a plurality of observation text sentences from a target environment, extracting one or more propositional logic values from the plurality of observation text sentences, finding a class for each propositional logic value by using external knowledge, converting each propositional logic value into a first-order logic by replacing a part in the propositional logic value with a variable word, the part indicating the class, selecting a LNN based on the class among LNNs prepared in advance for each class, each LNN receiving the one or more propositional logic values as a status input and outputting an action with a score indicating a degree of preference for taking the action, and performing a highest score action to the target environment to obtain a next state of the target environment and a reward for the highest score action.

Neuro-symbolic reinforcement learning with first-order logic

A perforated plate for a gas turbine combustor according to at least one embodiment is a perforated plate provided between a combustor basket and a combustor casing of the gas turbine combustor and fixed to an outer peripheral portion of the combustor basket. In a hole arrangement area with a plurality of through holes of the perforated plate, a region close to the combustor basket has a larger average value of a ligament ratio than a region close to the combustor casing, where the ligament ratio is obtained by dividing a distance between outer peripheral edges of two adjacent holes of the plurality of through holes by a distance between centers of the two holes.

Perforated plate for gas turbine combustor, gas turbine combustor, and gas turbine

A computer-implemented method is provided for machine prediction. The method includes forming, by a hardware processor, a Convolutional Dynamic Boltzmann Machine (C-DyBM) by extending a non-convolutional DyBM with a convolutional operation. The method further includes generating, by the hardware processor using the convolution operation of the C-DyBM, a prediction of a future event at time t from a past patch of time-series of observations. The method also includes performing, by the hardware processor, a physical action responsive to the prediction of the future event at time t.

Convolutional dynamic Boltzmann Machine for temporal event sequence

A method is provided for clustering data elements to extract specific patterns. The method specifies some data elements with a uniform distribution as a mini-batch and performs a single-pass cluster initialization by selecting a respective data element from the mini-batch as a respective initial cluster center to obtain cluster centers for clusters. The method assigns each data element in the mini-batch to a closest cluster by calculating a distance between each of the data elements in the mini-batch and each of the clusters. The method assigns k-minimum new centers by calculating an averaged distance to each data element in a same cluster. The method repeats the specifying step and the assigning steps responsive to a dissatisfaction of loop stop criteria which is based on distances between the centers and the K-minimum new centers. The method outputs a cluster id sequence responsive to a satisfaction of the loop stop criteria.

Mini-batch top-k-medoids for extracting specific patterns from CGM data

A light receiving unit includes photoelectric conversion elements () in which first pixels () are arrayed linearly along a long side of a parallelogram-shaped semiconductor substrate () and second pixels () are arrayed linearly at a location separated away from the first pixels () by a predetermined interpolation pixel distance. An output data processor that generates image data based on outputs of the photoelectric conversion elements () performs time correction of the image data by coupling image data based on outputs of the second pixels () with image data based on outputs of the first pixels () that have a time shift equal to a scanning time corresponding to the interpolation pixel distance.

Light receiving unit

There is provided a generating apparatus that generates an estimation model for estimating an attribute of an unknown target, the generating apparatus including a training-data acquiring unit acquiring training data including an attribute of a known target, attributes of other targets related to the known target, and a plurality of sets of types of relations between the known target and the other targets and a training processing unit training, using the training data, on the basis of the attribute of the known target and a type of a relation between the known target and the unknown target related to the known target, an estimation model for estimating the attribute of the unknown target, a generating method using the generating apparatus, and a program used in the generating apparatus.

Estimation model for estimating an attribute of an unknown customer

A solid-state imaging device includes blue photoelectric conversion elements, green photoelectric conversion elements, red photoelectric conversion elements, infrared photoelectric conversion elements, and an infrared cut filter (IRCF) is layered on the blue photoelectric conversion elements, the green photoelectric conversion elements, and the red photoelectric conversion elements with a uniform film thickness. Color filters that respectively transmit the blue light, the green light, and the red light are layered on the IRCF so as to correspond with the blue photoelectric conversion elements, the green photoelectric conversion elements, and the red photoelectric conversion elements. A visible-light shielding filter is layered on the infrared photoelectric conversion elements. An ultraviolet cut filter layered between the photoelectric conversion elements other than the infrared photoelectric conversion elements and the IRCF, and between the infrared photoelectric conversion elements and the visible-light shielding filter with a uniform film thickness.

Solid-state imaging device and image sensor for suppressing or preventing leaking of light into adjoining pixels

A computer-implemented method is presented for building a topic model to discover topics in a collection of documents generated by a plurality of users. The method includes extracting conversations from the collection of documents, dividing the extracted conversations into a plurality of segments, generating a topic distribution for each of the plurality of segments based on the extracted conversations and a first pre-defined prior probability distribution, and generating continuous value constructs for each of the topic distributions based on an external corpus and a second pre-defined prior probability distribution, wherein similarity is defined between the continuous value constructs.

Interactively building a topic model employing semantic similarity in a spoken dialog system

Text analysis includes determining one or more global analysis parameters based on backtranslation of a first corpus between supported languages. A new text analysis model is determined for an unsupported language based on the one or more global analysis parameters and a text analysis model for a first supported language. An input text is analyzed in the unsupported language with the new text analysis model.

Text analysis in unsupported languages using backtranslation

A method and system are provided for collecting social media users who have a specific profile. The method includes retrieving a set of lists connected by at least one criterion to a particular list that is included in a set of reliable lists whose users have already been reliably deemed to have a specific profile. The method includes calculating a list name based confidence value and a list member based confidence value for each list in the retrieved set of lists. The method includes updating the set of reliable lists by adding all lists in the retrieved set of lists that have the list name based confidence value above a first threshold value and the list member based confidence value above a second threshold value. The method includes outputting a listing of users belonging to set of reliable lists as the social media users who have the specific profile.

Collecting social media users in a specific customer segment

A method and system are provided. The method includes separating initial user-inputted data into terms. The method further includes determining which terms are identity-related terms relating to an identity of a user, and which terms are clarity-related terms relating to subject-matter clarity. The method also includes calculating a population. The population is a number of individuals satisfying the identity-related terms. The method additionally includes assigning, for each clarity-related term, a clarity level, representing a respective amount of clarity for a respective one of the clarity-related terms. The method further includes creating amended user-inputted data by amending the initial user-inputted data such that the population is greater than or equal to a first threshold, and such that the clarity level, assigned for each clarity-related term, is greater than or equal to a second threshold. The method also includes transmitting, using a graphical user interface, the amended user-inputted data to respondents.

Low privacy risk and high clarity social media support system

A plurality of characters including a player character appears in a video game. An arrangement position of at least one of characters is specified in a virtual space in accordance with progress of the video game. A display mode is set up to any one of a first display mode and a second display mode. A game control is carried out in the first display mode in which a photographed image of the virtual space photographed by a virtual camera is displayed on a game screen or the second display mode in which an arrangement image is displayed on the game screen. Each of the plurality of characters is arranged at a predetermined position in the arrangement image with a predetermined relationship with the arrangement position of the corresponding character in the virtual space. The game control is carried out in the display mode thus set up.

Video game processing program and video game processing method

A plurality of characters including a player character appears in a video game. An arrangement position of at least one of characters is specified in a virtual space in accordance with progress of the video game. A battle mode is set up to any one of a first battle mode and a second battle mode. A battle control is carried out in the first battle mode in which a photographed image of the virtual space photographed by a virtual camera is displayed on a battle screen or the second battle mode in which an arrangement image is displayed on the battle screen. Each of the plurality of characters is arranged at a predetermined position in the arrangement image with no relationship with the arrangement position of the corresponding character in the virtual space. The battle control is carried out in the battle mode thus set up.

The characteristics of a heat radiation member used in a semiconductor module are improved. A heat radiation member (A) including an aluminum type member () which contains aluminum and a copper type member () which contains copper, which is embedded in the aluminum type member (), and sides of which are enclosed by the aluminum type member () is formed. A semiconductor element is thermally bonded to the heat radiation member (A) to fabricate a semiconductor module. The heat radiation member (A) includes the aluminum type member () and the copper type member (). As a result, it is possible to realize light weight while ensuring certain heat radiation. In addition, the copper type member () is enclosed by the aluminum type member (). Accordingly, the strength of the heat radiation member (A) can be increased.

Semiconductor module and heat radiation member

An acoustic damper is fixed to a vibration source to be provided along an outer surface of the vibration source, forms a passage that takes in air vibration generated by the vibration source, and is provided with an acoustic portion having a partition member closing the downstream side of the passage with respect to the propagation of the air vibration to serve as resistance against the air vibration, in which the acoustic portion allows air vibration to propagate in a reverse direction, and has partition members arranged so as to serve as resistance against the air vibration in respective directions. This allows to prevent from generating stress due to pressure fluctuation caused when reducing vibration, and to achieve a relatively compact size.

Acoustic damper, combustor, and gas turbine

Growing community of inventors

Takasago, Japan

Masaki Ono