<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <title>Repository Collection: null</title>
  <link rel="alternate" href="https://scholar.dgist.ac.kr/handle/20.500.11750/6304" />
  <subtitle />
  <id>https://scholar.dgist.ac.kr/handle/20.500.11750/6304</id>
  <updated>2026-04-04T19:43:20Z</updated>
  <dc:date>2026-04-04T19:43:20Z</dc:date>
  <entry>
    <title>Hardware accelerator for performing computation of deep neural network and electronic device including same</title>
    <link rel="alternate" href="https://scholar.dgist.ac.kr/handle/20.500.11750/58824" />
    <author>
      <name>노석환</name>
    </author>
    <author>
      <name>궁재하</name>
    </author>
    <author>
      <name>구자현</name>
    </author>
    <id>https://scholar.dgist.ac.kr/handle/20.500.11750/58824</id>
    <updated>2025-12-30T01:47:48Z</updated>
    <summary type="text">Title: Hardware accelerator for performing computation of deep neural network and electronic device including same
Author(s): 노석환; 궁재하; 구자현
Abstract: A hardware accelerator includes a processing core including a plurality of multipliers configured to perform one-dimensional (1D) sub-word parallelization between symbols and mantissas of a first tensor and symbols and mantissas of a second tensor, a first processing device configured to operate in a two-dimensional (2D) mode of operation in which the first tensor and the second tensor are coupled to each other, and a second processing device configured to operate in a two-dimensional (2D) mode of operation in which the first tensor and the second tensor are coupled to each other. And a second processing device configured to operate in a three-dimensional (3D) operation mode in which the calculation results of the plurality of multipliers are accumulated in a channel direction, and then a result of accumulating the calculation results is output.</summary>
  </entry>
  <entry>
    <title>신경망 모델에 기반하여 고주파 생체 신호를 복원하는 방법 및 장치</title>
    <link rel="alternate" href="https://scholar.dgist.ac.kr/handle/20.500.11750/58353" />
    <author>
      <name>강홍기</name>
    </author>
    <author>
      <name>홍나리</name>
    </author>
    <author>
      <name>진경환</name>
    </author>
    <author>
      <name>이정협</name>
    </author>
    <author>
      <name>궁재하</name>
    </author>
    <author>
      <name>이재원</name>
    </author>
    <id>https://scholar.dgist.ac.kr/handle/20.500.11750/58353</id>
    <updated>2025-07-25T04:08:29Z</updated>
    <summary type="text">Title: 신경망 모델에 기반하여 고주파 생체 신호를 복원하는 방법 및 장치
Author(s): 강홍기; 홍나리; 진경환; 이정협; 궁재하; 이재원
Abstract: 본 개시의 일 실시 예에 따른 고주파 생체 신호를 복원하는 방법은 프로세서에 의해, 제1 생체 신호를 로딩하는 단계 및 저주파 신호인 제1 생체 신호를 제1 신경망 모델에 기반하여 고주파 신호인 제2 생체 신호로 변환하는 단계를 포함할 수 있다.</summary>
  </entry>
  <entry>
    <title>완화된 프루닝을 통한 행렬 데이터 처리 방법 및 그 장치</title>
    <link rel="alternate" href="https://scholar.dgist.ac.kr/handle/20.500.11750/58352" />
    <author>
      <name>궁재하</name>
    </author>
    <author>
      <name>박준기</name>
    </author>
    <author>
      <name>김재준</name>
    </author>
    <id>https://scholar.dgist.ac.kr/handle/20.500.11750/58352</id>
    <updated>2025-07-25T03:22:55Z</updated>
    <summary type="text">Title: 완화된 프루닝을 통한 행렬 데이터 처리 방법 및 그 장치
Author(s): 궁재하; 박준기; 김재준</summary>
  </entry>
</feed>

