Nippon Telegraph and Telephone Corporation

International Business Machines Corporation

Subaru Corporation

Buresuto Kogyo Co., Ltd.

Junichi Kato

Kenji Kawai

Takeshi Sakamoto

Yuki Arikawa

Masaharu Sakamoto

Tsuyoshi Ito

Takayuki Kushida

Tomoko Murayama

Kazuto Yamafuji

Huycu Ngo

Nobuyuki Tanaka

Tomoaki Kawamura

Yutaka Oishi

Satoshi Shigematsu

Yoshinori Tahara

Hiroyuki Uzawa

Kumiko Maeda

Namiko Ikeda

Yuriko Nishikawa

Chiaki Oishi

Saki Hatta

Katsuhiko Hagiwara

Shoko Ohteru

Naoki Miura

Yosuke Kubota

Kazuhiko Terada

Taiki Sakai

Yoshihiro Nakazato

Nobusuke Nagaoka

Keiichi Sugaya

A distributed deep learning system that can achieve speeding-up by processing learning in parallel at a large number of learning nodes connected with a communication network and perform faster cooperative processing among the learning nodes connected through the communication network is provided. The distributed deep learning system includes: a plurality of computing interconnect devicesconnected with each other through a ring communication networkthrough which communication is possible in one direction; and a plurality of learning nodesconnected with the respective computing interconnect devicesin a one-to-one relation, and each computing interconnect deviceexecutes communication packet transmission-reception processing between the learning nodesand All-reduce processing simultaneously in parallel.

Distributed deep learning system, distributed deep learning method, and computing interconnect device

Each of learning nodes calculates gradients of a loss function from an output result obtained by inputting learning data to a learning target neural network, converts a calculation result into a packet, and transmits the packet to a computing interconnect device. The computing interconnect device receives the packet transmitted from each of the learning nodes, acquires a value of the gradients stored in the packet, calculates a sum of the gradients, converts a calculation result into a packet, and transmits the packet to each of the learning nodes. Each of the learning nodes receives the packet transmitted from the computing interconnect device and updates a constituent parameter of a neural network based on a value stored in the packet.

Distributed deep learning system using a communication network for stochastic gradient descent calculations

Individual distributed processing nodes packetize distributed data for each weight of a neural network of a learning object in an order of a number of the weight, transmit the distributed data to an aggregation processing node, acquire aggregation data transmitted from the node in order, and update the weight of the neural network. The node acquires the transmitted distributed data, packetizes the aggregation data for which the distributed data of all the distributed processing nodes is aggregated for each weight, and transmits the aggregation data to the individual nodes. The individual nodes monitor an unreceived data amount which is a difference between data amounts of the transmitted distributed data and the acquired aggregation data, and when the unreceived data amount becomes equal to or larger than a threshold Ma, stops transmission of the distributed data until the unreceived data amount becomes equal to or smaller than a threshold Mb (Mb&#x3c;Ma).

Distributed processing system and distributed processing method

A first distributed processing node transmits distributed data to a second distributed processing node as intermediate consolidated data. A third distributed processing node generates intermediate consolidated data after update from received intermediate consolidated data and distributed data, and transmits the intermediate consolidated data to a fourth distributed processing node. The first distributed processing node transmits the received intermediate consolidated data to fifth distributed processing node as consolidated data. The third distributed processing node transmits the received consolidated data to a sixth distributed processing node. When an aggregation communication time period required by each distributed processing node to consolidate the distributed data or an aggregation dispatch communication time period being a total time period of the aggregation communication time period and a time period required by each distributed processing node to dispatch the consolidated data exceeds a predetermined time period, the first distributed processing node issues a warning.

A vehicle control apparatus includes an inverter controller. The inverter controller holds a plurality of control maps for an inverter. The inverter supplies electric power to a drive motor. The drive motor drives a drive wheel of the vehicle. The inverter controller selects any one of the plurality of control maps on a basis of a notification instruction that instructs to notify a driver of information. The inverter controller controls an operation of the inverter on a basis of the control map selected.

Vehicle control apparatus

An OLT () is provided with a priority control bypass circuit () and an encryption/decryption bypass circuit (), or an ONU () is provided with a priority control bypass circuit () and an encryption/decryption bypass circuit (), and one or both of encryption/decryption processing and priority control processing are bypassed in accordance with a priority control bypass instruction (BP) and an encryption/decryption bypass instruction (BE), which are set in advance. This reduces a processing delay that occurs in the OLT or the ONU.

PON system, wireless network system, data transfer method, OLT, and ONU

A method, a computing system and a computer program product are provided. A link for use by a user to access a file is created. Content of the file is encrypted using a common key. The common key is encrypted using a public key of the user and is registered in the link. Access rights regarding the file are set for the user and registered in the link. The link includes information for use by the user to access the file when the access rights indicate that the user is authorized to access the file.

File sharing and policy control based on file link mechanism

A technique of operating a system is provided that processes data with one or more data processing modules provided in parallel. A synchronization token is input into at least one data processing module that is in an operational state from the one or more data processing modules provided in parallel, in response to a request to change allocation of the data. The allocation of the data is changed to the one or more data processing modules provided in parallel, after the synchronization token is input. In response to the synchronization token having arrived at a data processing module at a later stage, the at least one data processing module, that stopped the processing, starts processing data after the synchronization token is input to the at least one data processing module.

Method for changing allocation of data using synchronization token

The present invention may be a method, a computer system, and a computer program product. An embodiment of the present invention provides a method for judging data consistency in a database. In one embodiment, the method comprises the following: generating a property of data obtained from a first database; associating the property with an attribute of a data model to generate a data property definition; judging whether data obtained from a second database satisfies the data property definition or not; and outputting a result of the judgment. In another embodiment, the method comprises the following: generating a property of data obtained from a database; associating the property with an attribute of a data model to generate a data property definition; judging whether data which is stored in the database satisfies the data property definition or not; and outputting a result of the judgment.

Judgement of data consistency in a database

The present invention may be a method, a computer system, and a computer program product. An embodiment of the present invention provides a method for finding a problem in procedures described in a guide document for install and configuration of software. The method comprises calculating, using a dynamic programming matching, a distance between an install-and-configuration log generated by executing the install and configuration of the software according to the guide document at a user-side computer and a log template generated by executing the install and configuration of the software according to the guide document at an administrator-side computer, and finding a problem in the procedures, using the distance.

Finding a problem in procedures described in a guide document

An upstream allocation circuit () and a downstream allocation circuit () are provided in an OLT (). For example, a superimposed frame obtained by bundling upstream frames (upstream control frames+upstream data frames) from all ONUS is input to the upstream allocation circuit () via a frame reproduction circuit (-). The superimposed frame may be generated at the stage of optical signals or generated after converting optical signals into electrical signals. The upstream allocation circuit () allocates each of the upstream control frames bundled into the superimposed frame to a predetermined PON control circuit () based on information (PON port number or LLID) added to the frames. The downstream allocation circuit () allocates, to a preset frame reproduction circuit (), each downstream control frames output from the PON control circuits ().

Station-side apparatus in optical transmission system and optical transmission system

A selection and distribution circuit () is provided between N optical transceivers () and one PON control circuit (). The selection and distribution circuit () selects the optical transceiver () corresponding to an upstream frame that time-divisionally arrives, thereby transferring the upstream frame opto-electrically converted by the transceiver () to the PON control circuit () and distributing a downstream frame from the PON control circuit () to each optical transceiver (). A power supply control circuit () stops power supply to at least one of one of optical transceivers () that are not used to transfer the frame of the optical transceivers () and a circuit that is not used to transfer the frame in the selection and distribution circuit (). This can reduce the system cost per ONU in the optical transmission system.

Growing community of inventors

Tokyo, Japan

Junichi Kato