公开(公告)号：US20240069875A1

公开(公告)日：2024-02-29

申请号：US18209617

申请日：2023-06-14

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Jae-Bok PARK ,  Chang-Sik CHO ,  Kyung-Hee LEE ,  Ji-Young KWAK ,  Seon-Tae KIM ,  Hong-Soog KIM ,  Jin-Wuk SEOK ,  Hyun-Woo CHO

IPC: G06F8/36 ,  G06F8/60 ,  G06F9/451

CPC classification number: G06F8/36 ,  G06F8/60 ,  G06F9/453

Abstract: Disclosed herein are a neural network model deployment method and apparatus for providing a deep learning service. The neural network model deployment method may include providing a specification wizard to a user, searching for and training a neural network based on a user requirement specification that is input through the specification wizard, generating a neural network template code based on the user requirement specification and the trained neural network, converting the trained neural network into a deployment neural network that is usable in a target device based on the user requirement specification, and deploying the deployment neural network to the target device.

公开(公告)号：US20220245458A1

公开(公告)日：2022-08-04

申请号：US17485322

申请日：2021-09-24

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Jae-Bok PARK

IPC: G06N3/08 ,  G06K9/62 ,  G06K9/72

Abstract: Disclosed herein are an apparatus and method for converting a neural network. The method includes separating neural network data of a source framework to form a tree structure by analyzing the same, converting the neural network data in a tree structure to a neural network optimized for a target framework, classifying training data based on the result of analysis of the neural network data of the source framework, converting the classified training data to the training data structure of the target framework, and creating a neural network and training data of the target framework by combining the converted neural network and the converted training data.

公开(公告)号：US20230315402A1

公开(公告)日：2023-10-05

申请号：US18098402

申请日：2023-01-18

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Chang-Sik CHO ,  Jae-Bok PARK ,  Kyung-Hee LEE ,  Ji-Young KWAK ,  Seon-Tae KIM ,  Ik-Soo SHIN

IPC: G06F8/35 ,  G06N3/10 ,  G06F8/10

CPC classification number: G06F8/35 ,  G06N3/10 ,  G06F8/10

Abstract: Disclosed herein are an apparatus and method for developing a neural network application. The apparatus includes one or more processors and executable memory for storing at least one program executed by the one or more processors. The at least one program receives a target specification and an application specification including user requirements, searches for a neural network model corresponding to the target specification and the application specification in a database, builds an inference engine for performing a neural network operation used by the neural network model, and generates a target image for executing the neural network model to be suitable for a target device using the inference engine.

公开(公告)号：US20220374740A1

公开(公告)日：2022-11-24

申请号：US17767364

申请日：2020-09-28

Applicant: Electronics and Telecommunications Research Institute

Inventor： Chang-Sik CHO ,  Jae-Bok PARK ,  Seung-Mok YOO ,  Seok-Jin YOON ,  Kyung-Hee LEE

IPC: G06N5/04

Abstract: An embodiment relates to an artificial intelligence inference apparatus and method. The embodiment provides an artificial intelligence inference method, and may include converting an application based on a previously learned neural network into executable code in a high-level language independent of a learning framework, separating the executable code into General-Purpose Language (GPL) code and Domain-Specific Language (DSL) code depending on whether an acceleration operation is required, and generating target code optimized for hardware from the separated GPL code and DSL code.

公开(公告)号：US20230342133A1

公开(公告)日：2023-10-26

申请号：US18098486

申请日：2023-01-18

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Kyung-Hee LEE ,  Ji-Young KWAK ,  Seon-Tae KIM ,  Jae-Bok PARK ,  Ik-Soo SHIN ,  Chang-Sik CHO

IPC: G06F8/61 ,  G06F8/10

CPC classification number: G06F8/63 ,  G06F8/10

Abstract: Disclosed herein are an apparatus and method for generating a neural network executable image. The apparatus includes one or more processors and executable memory for storing at least one program executed by the one or more processors. The at least one program receives user requirements including a default neural network model and training result data for generating a neural network executable image required by a user, checks whether the default neural network model included in the user requirements is capable of being supported in a target system in which the neural network executable image is to be installed, converts the default neural network model into a neural network model executable in the target system, converts the training result data by reconfiguring the data format set of the training result data, and generates a neural network executable image by combining the converted neural network model and the converted training result data.

公开(公告)号：US20230316091A1

公开(公告)日：2023-10-05

申请号：US18166948

申请日：2023-02-09

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Jin-Wuk SEOK ,  Ji-Young KWAK ,  Seon-Tae KIM ,  Hong-Soog KIM ,  Jae-Bok PARK ,  Kyung-Hee LEE ,  Chang-Sik CHO ,  Hyun-Woo CHO

IPC: G06N3/098

CPC classification number: G06N3/098

Abstract: Disclosed herein are a federated learning method and apparatus. The federated learning method includes receiving a feature vector extracted from a client side and label data corresponding to the feature vector, outputting a feature vector with phase information preserved therein by applying the feature vector as input of a Self-Organizing Feature Map (SOFM), and training a neural network model by applying both the feature vector with the phase information preserved therein and the label data as input of a neural network model.