MEMBER
Gunho Park
Ph.D. Student,
Department of Electrical Engineering,
Pohang University of Science and Technology (POSTECH)
Contact Information
Address : #308, Science Building II, POSTECH, Pohang, Republic of Korea
Email : gunho1123[at]postech.ac.kr
Office : +82-(0)54-279-5587
Educational Backgrounds
[2021/02~Present] Ph.D. Candidate in Electrical Engineering, POSTECH, Pohang, Republic of Korea
[2019/02~2021/02] M.S. in Electrical Engineering, POSTECH, Pohang, Republic of Korea
[2012/03~2019/02] B.S. in Electrical Engineering, POSTECH, Pohang, Republic of Korea
Professional Experiences
[2023/09~Present] Research Intern, NAVER CLOUD, Seongnam, Republic of Korea
[2018/01~Present] Research Assistant, POSTECH, Pohang, Republic of Korea
[2021/09~2022/03] Research Intern, NAVER CLOVA, Seongnam, Republic of Korea
[2018/12~2019/02] Software Engineering Intern, SK Hynix, Icheon, Republic of Korea
[2017/09~2017/12] Exchange Scholarship, Univ. of California, Berkeley, CA, United States
[2017/01~2017/12] Backend Developer Intern, byclip, Seoul, Republic of Korea
Research Interests
Intelligent Mobile System-on-Chip
ASIC/FPGA Implementation
General Purpose Graphic Processing Unit (GPGPU)
Approximate Computing
Deep Learning
Publications
International Journal Papers
Gunho Park, Jaeha Kung, and Youngjoo Lee*, "Simplified compressor and encoder designs for low-cost approximate radix-4 Booth multiplier," IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 70, no.3, pp. 1154-1158, Mar. 2023.
Gunho Park, Jaeha Kung, and Youngjoo Lee*, "Design and analysis of approximate compressors for balanced error accumulation in MAC operator," IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 68, no. 7, pp. 2950-2961, July 2021.
International Conference Proceedings
Gunho Park+, Baeseong Park+, Minsub Kim, Sungjae Lee, Jeonghoon Kim, Beomseok Kwon, Se Jung Kwon, Byeongwook Kim, Youngjoo Lee, and Dongsoo Lee*, "LUT-GEMM: Quantized matrix multiplication based on LUTs for efficient inference in large-scale generative language models," International Conference on Learning Representations (ICLR), Vienna, Austria, May 2024.
Jiwoo Kim, Gunho Park, and Youngjoo Lee*, "Low-power encoder and compressor design for approximate radix-8 Booth multiplier," IEEE International Symposium on Circuits and Systems (ISCAS), Singapore, Singapore, May 2024.
Jung Gyu Min, Dongyun Kam, Younghoon Byun, Gunho Park, and Youngjoo Lee*, "Energy-efficient RISC-V-based vector processor for cache-aware structurally-pruned transformers," ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED), Vienna, Austria, Aug. 2023.
Eunji Yoo+, Gunho Park+, Jung Gyu Min, Se Jung Kwon, Baeseong park, Dongsoo Lee, and Youngjoo Lee*, "TF-MVP: Novel sparsity-aware transformer accelerator with mixed-length vector pruning," Design Automation Conference (DAC), San Francisco, CA, USA, July 2023.
Younghoon Byun+, Seungsik Moon+, Baeseong Park, Se Jung Kwon, Dongsoo Lee, Gunho Park, Eunji Yoo, Jung Gyu Min, and Youngjoo Lee*, "Sparsity-aware memory interface architecture using stacked XORNet compression for accelerating pruned-DNN models," Conference on Machine Learning and Systems (MLSys), Miami, FL, USA, June 2023.
Domestic Conference Proceedings
심도은, 박군호, 이영주*, "Low-precision posit과 floating-point arithmetic의 정밀도 및 하드웨어 복잡도 분석," 전자공학회 하계종합학술대회, June 2022.
박군호, 궁재하, 김영석, 김미정, 이영주*, "저전력 딥러닝 가속기를 위한 근사 곱셈기 설계 프레임워크," 대한전자공학회 하계종합학술대회, July 2021. (네이버 논문상)
박군호, 윤형석, 이재호, 김영석, 김미정, 이영주*, "FPGA 기반의 고해상도 generative adversarial network 가속기 구현," 제 26회 한국반도체학술대회, Feb. 2019, pp. 825.
이재호, 윤형석, 박군호, 김영석, 김미정, 이영주*, "Super resolution generative adversarial network (SRGAN) 모델 경량화," 제 26회 한국반도체학술대회, Feb. 2019, pp. 826. (Best Poster Award)
Patents
이영주, 박군호, 김영석, "부분곱 생성기, 부분곱 누적기 및 이들을 포함하는 근사 곱셈기," 출원번호: 10-2022-0160911.
Youngjoo Lee, Young-Seok Kim, Hyunhoon Lee, and Gunho Park, "Method and device for high-speed image recognition using 3D CNN," US17/422,161.
Youngjoo Lee, Young-Seok Kim, Hyunhoon Lee, and Gunho Park, "Method and apparatus for high-speed image recognition using 3D convolutional neural network," PCT/KR2020/000647.
이영주, 김영석, 박군호, 이현훈, "동적으로 3D CNN을 이용하는 고속 영상 인식 방법 및 장치," 등록번호: 10-2263005, 출원번호: 10-2019-0005200.
이영주, 김영석, 박군호, 이현훈, "3D CNN을 이용한 고속 영상 인식 방법 및 장치," 등록번호: 10-2263017, 출원번호: 10-2019-0005188.
Open Source Projects
TensorFlow Model Developer - Google x Postech
BASNet: Boundary-Aware Salient Object Detection (link)
DETR: End-to-End Object Detection with Transformers (link)
Pix2Seq: A Language Modeling Framework for Object Detection (link)
Awards and Honors
Best Paper Award (네이버 논문상), Summer Annual Conference of IEIE, 2021.
Best Poster Award, 제 26회 한국반도체학술대회, 2019.
Excellence Award, Intel FPGA Design Contest, 2018.