I have designed and mass-produced semiconductors for 10 years since 2006.
In particular, I was in charge of hardware accelerator and full chip design in the field of display driver IC and touch screen controllers.
And I have been working in the field of artificial intelligence since 2017.
I designed NPU architecture by the end of 2020 and worked on performance modeling of NLP model by the end of February 2022.
I am currently in charge of the chip design automation.
Chip Design Automation
Optimizations
Mar. 2021 ~ Feb. 2023: MS student in Department of Computer Science and Engineering, Seoul National University (Advisor: Dr. Jaejin Lee)
Mar. 1997 ~ Feb. 2006: BS student in School of Electrical Engineering and Computer Science, Kyungpook National University
Samsung Advanced Institute of Technology (2019~)
Machine Learning TU(2022~)
Chip design automation
Leader of Physical Design Activity
1D/2D placement optimization project collaborating with DRAM Design Team
Green Supercomputer Architecture, Computing Platform Lab, System Center (2021~2022)
Analytic performance model design for NLP model and miniDFT
GPU power efficiency enhancement for HPL
Hyperscale Computing Platform, Computing Platform Lab, AI & S/W Center (2019~2020)
High utilization MAC architecture design and evaluation
High-level simulator design for the PPA analysis of the MAC architecture
Core architecture design for inference
Feasibility evaluation for enhanced lossless compression algorithm
Winograd algorithm feasibility evaluation
Samsung Electronics Semiconductor Research Center (2014~2018)
Future Technology & Convergence Research Project TF (2018)
Future technology investigation and future project design
Feasibility evaluation for embedded FPGA
Deep Learning Platform Project in C-Lab (2017)
Team leader
2D systolic array-based NPU using FPGA
New Lossless compression algorithm design
SW R&D Center (2014~2016)
Device driver design for in-house secure OS
Tool development for RTL design automation
Samsung Electronics Display Solution Team (2006~2014)
LCD low power algorithm, IP design, FPGA evaluation, and mass production (Galaxy, iPhone, etc.)
OLED low power algorithm and IP design
Dynamic Capacitance Compensation IP design and mass production (Samsung digital camera)
World’s first Touch + Display Driver IC chip design
Touch Screen Controller full chip design, H/W accelerator IP design, and mass production (Samsung Galaxy S7~, OPPO and VIVO, etc.)
Samsung S/W membership (2005)
Leader of SoC Team
5-stage pipelined MIPS CPU design using FPGA
Kyungpook National University (1997~2006)
Robot arm imitating hand movements (2004)
Sign language translator and chat system (2005)
[Thesis] Hyungdal Kwon, Jaejin Lee, "Greedy-K: A New Local Optimization Algorithm for Solving 1D Circuit Placement Problems"
Jinpyo Kim, Hyungdal Kwon, Jintaek Kang, Jihwan Park, Seungwook Lee, Jaejin Lee. "SnuHPL: high performance LINPACK for heterogeneous GPUs" Proceedings of the 36th ACM International Conference on Supercomputing (ICS), 2022
Donghyun Kang, Jintaek Kang, Hyungdal Kwon, Hyunsik Park, Soonhoi Ha. "A Novel Convolutional Neural Network Accelerator That Enables Fully-Pipelined Execution of Layers" IEEE 37th International Conference on Computer Design (ICCD), 2019
Ki-Duk Kim, San-Ho Byun, Yoon-Kyung Choi, Jong-Hak Baek, Hwa-Hyun Cho, Jong-Kang Park, Hae-Yong Ahn, Chang-Ju Lee, Min-Soo Cho, Joo-Hyeon Lee, Sang-Woo Kim, Hyungdal Kwon, Yong-Yeob Choi, Chang-Ju Lee, Hwa-Hyun Cho, Jae-Suk Yu, Myunghee Lee. "A capacitive touch controller robust to display noise for ultrathin touch screen displays" IEEE International Solid-State Circuits Conference, 2012
Hyoung-Rae Kim, Yoon-Kyung Choi, San-Ho Byun, Sang-Woo Kim, Kwang-Ho Choi, Hae-Yong Ahn, Jong-Kang Park, Dong-Yul Lee, Zhong-Yuan Wu, Hyung-Dal Kwon, Yong-Yeob Choi, Chang-Ju Lee, Hwa-Hyun Cho, Jae-Suk Yu. "A mobile-display-driver IC embedding a capacitive-touch-screen controller system" IEEE International Solid-State Circuits Conference - (ISSCC), 2010
Method and apparatus with performance modeling (US20240061972A1)
Method and apparatus with deep learning operations (US17/356771)
Method and apparatus for performing deep learning operations (US17/102884)
Method and apparatus with neural network data input and output control (US16/900007)
Operating method of power optimization scheduler and computing apparatus including power optimization scheduler (US20210247838A1)
Method and apparatus for performing deep learning operations (US17/168457)
Memory device and method (US20210089610A1)
Neural processing unit, neural processing system, and application system (US20200151549A1)
Electronic system including FPGA and operation method thereof (US11012075)
Method and device for controlling data input and output of fully connected network (US16/970859)
Touch screen controller for increasing data processing speed and touch system including the same (US10355737, US10141972)
Value adjustment methods, value adjustment signal processing apparatus, and image display systems using the same (US8791953)
완전 연결 네트워크의 데이터 입력 및 출력을 제어하는 방법 및 장치 (P1020180020014)
디스플레이 제어기 및 디스플레이 제어 방법 (KR20130084765A)
디스플레이 패널 제어 장치 및 디스플레이 패널의 제어 프로그램이 기록된 컴퓨터로 판독 가 능한 기록 매체 (P1020170027638)
부동산 정보 제공 및 공유 장치 (KR101692521)
부동산 거래 가치 관리 장치 (KR101691305)
무한탐구상(Boundless Search for Breakthroughs Award, Jan. 2022)
Green500 2nd/Top500 11th (Nov. 2021)
Best Achievement Award (Jan. 2021)
Prize of Business team leader of DDI (LCD low power algorithm IP) (2007)
Joint Grand Prize at IDEC Design Competition. (MIPS CPU) (2005)