公开(公告)号：US20230087752A1

公开(公告)日：2023-03-23

申请号：US17910270

申请日：2020-03-26

Applicant: NEC Corporation

Inventor： Yuki KOBAYASHI

IPC: G06N3/084

Abstract: An information processing device 500 of the present invention includes a first learning means 521 for performing a first learning process by using a first learning data set to generate a first learning model including a first weight parameter, a second learning means 522 for performing a second learning process by using a second learning data set to generate a second learning model including a second weight parameter, and an arbitration means 523 for, when performing the first learning process and the second learning process, updating the first weight parameter and the second weight parameter such that the values of the first weight parameter and the second weight parameter become almost the same.

公开(公告)号：US20240403645A1

公开(公告)日：2024-12-05

申请号：US18667147

申请日：2024-05-17

Applicant: NEC Corporation

Inventor： Yuki KOBAYASHI

IPC: G06N3/082

Abstract: The neural network structure proposal device facilitates finding a neural network structure suitable for a device includes an operational efficiency analysis unit which calculate, for each of a plurality of layers of a neural network having different parameters, an estimated amount of execution time of the layer and operational efficiency corresponding to the computation amount per unit time on a target device, and a layer structure replacing unit which replaces the layer with the large estimated amount of execution time and low operational efficiency with another layer, and outputs neural network structure information indicating structure of the neural network.

公开(公告)号：US20230359691A1

公开(公告)日：2023-11-09

申请号：US18023436

申请日：2020-09-03

Applicant: NEC Corporation

Inventor： Takuya ARAKI ,  Fumiyo TAKANO ,  Hiroaki INOUE ,  Motoi SUZUKI ,  Yuki KOBAYASHI ,  Takahiro NISHIMURA ,  Hiroshi CHISHIMA ,  Akihiro YATABE

IPC: G06F17/11

CPC classification number: G06F17/11

Abstract: The conversion means 72 generates a program for deriving a QUBO matrix that is a matrix that defines an energy function of QUBO, by converting a description of a natural language representing a combinatorial optimization problem into a description of the program for deriving the QUBO matrix. It is possible to generate a program for deriving a QUBO matrix based on a description of a natural language representing a combinatorial optimization problem.

公开(公告)号：US20240037185A1

公开(公告)日：2024-02-01

申请号：US18011329

申请日：2020-06-29

Applicant: NEC Corporation

Inventor： Fumiyo TAKANO ,  Hiroaki INOUE ,  Takuya ARAKI ,  Motoi SUZUKI ,  Yuki KOBAYASHI ,  Takahiro NISHIMURA ,  Hiroshi CHISHIMA ,  Akihiro YATABE

IPC: G06F17/18

CPC classification number: G06F17/18

Abstract: The inconsistency determination device 1 includes inconsistency determination means 9. The inconsistency determination means 9 determines whether or not there is inconsistency in an expression, which is to be converted into an energy function of a model representing states of individual spins by a first value or a second value, representing energy in a combinatorial optimization problem.

公开(公告)号：US20230306074A1

公开(公告)日：2023-09-28

申请号：US18020526

申请日：2020-08-20

Applicant: NEC Corporation

Inventor： Motoi SUZUKI ,  Hiroaki INOUE ,  Takuya ARAKI ,  Fumiyo TAKANO ,  Yuki KOBAYASHI ,  Hiroshi CHISHIMA ,  Takahiro NISHIMURA ,  Akihiro YATABE

IPC: G06F17/11 ,  G06F17/16

CPC classification number: G06F17/11 ,  G06F17/16

Abstract: The difference matrix calculation means 72 calculates a difference matrix which is a difference between a first matrix representing a combinatorial optimization problem for which a solution was found last time and a second matrix representing the latest combinatorial optimization problem. The difference matrix compression means 73 compresses the difference matrix. The transmission means 74 transmits the difference matrix after compression to a combinatorial optimization problem solution device that finds a solution to the combinatorial optimization problem.

公开(公告)号：US20210256099A1

公开(公告)日：2021-08-19

申请号：US17271270

申请日：2019-12-03

Applicant: NEC Corporation

Inventor： Tomohiro MIWA ,  Joji TANAKA ,  Yoshikazu ARAI ,  Keiji KANAMEDA ,  Tsuyoshi NAKAMURA ,  Yuki KOBAYASHI

IPC: G06F21/31 ,  G06K9/00 ,  G06K9/62

Abstract: An information processing method includes acquiring image information imaged in such a manner that identification information for identifying a user and a scene related to the user are associated with each other (step S11), and performing a process of associating the identification information registered in advance with the image information including the scene related to a user who is the same as the user identified by the identification information (step S12).

公开(公告)号：US20250124731A1

公开(公告)日：2025-04-17

申请号：US18903087

申请日：2024-10-01

Applicant: NEC Corporation

Inventor： Yuki KOBAYASHI ,  Tsuyoshi NAKAMURA ,  Yoshikazu ARAI ,  Kenichi UEDA ,  Shinnosuke NiSHIMOTO

IPC: G06V30/18 ,  G06F16/33

Abstract: The acquisition means acquires reader information and content information. The prompt generation means generates a prompt based on the reader information and the content information. The content editing means inputs the prompt into a large language model and acquire an output from the large language model as edited content.

公开(公告)号：US20240202496A1

公开(公告)日：2024-06-20

申请号：US18286304

申请日：2021-04-20

Applicant: NEC Corporation

Inventor： Yuki KOBAYASHI

IPC: G06N3/045 ,  G06N3/0895

CPC classification number: G06N3/045 ,  G06N3/0895

Abstract: A neural network structure search device includes a training unit which trains a neural network model with a first neural network structure using a data set for training, a first generation unit which generates analysis information indicating an importance of each of elements comprising the first neural network structure by analyzing a trained model generated from the neural network model by the training using the data set for training, an identifying unit which identifies an element in the first neural network structure whose importance is lower than a predetermined value using the generated analysis information, and a second generation unit which generates a second neural network structure based on the first neural network structure by deleting the identified element from the first neural network structure.

公开(公告)号：US20230359921A1

公开(公告)日：2023-11-09

申请号：US18026056

申请日：2021-08-31

Applicant: NEC Corporation

Inventor： Takahiro NISHIMURA ,  Hiroaki INOUE ,  Takuya ARAKI ,  Motoi SUZUKI ,  Fumiyo TAKANO ,  Yuki KOBAYASHI ,  Hiroshi CHISHIMA ,  Akihiro YATABE

IPC: G06N10/60 ,  G06F17/11

CPC classification number: G06N10/60 ,  G06F17/11

Abstract: An information processing device that compares a data amount of a QUBO model to be processed by an annealing engine and a data amount of a program representing the QUBO model, and that transmits whichever has a lower data amount to the annealing engine.

公开(公告)号：US20230315807A1

公开(公告)日：2023-10-05

申请号：US18015246

申请日：2020-07-17

Applicant: NEC Corporation

Inventor： Fumiyo TAKANO ,  Hiroaki INOUE ,  Takuya ARAKI ,  Motoi SUZUKI ,  Yuki KOBAYASHI ,  Hiroshi CHISHIMA ,  Akihir YATABE ,  Takahiro NISHIMURA

IPC: G06F17/18

CPC classification number: G06F17/18

Abstract: The annealing execution means 73 executes annealing using an energy function, when a solution to a combinatorial optimization problem, and the energy function of a model representing states of individual spins by a first value or a second value which is used to obtain the solution are given, and derives multiple new solutions to the combinatorial optimization problem. The solution determination means 74 determines a solution from among the multiple solutions derived by the annealing execution means 73, based on a solution determination rule for determining a solution among the multiple new solutions and the given solution to the combinatorial optimization problem.