Seoul National University's (SNU’s) College of Engineering announced on May 8th that Professor Yongtaek Hong from the Department of Electrical and Computer Engineering (ECE) has been selected as a Fellow of the Society for Information Display (SID) this year. Professor Hong is recognized as a leading expert in the field of stretchable display technology. He developed technologies that integrate transistors, light-emitting devices, sensors, and driving chips on stretchable substrates using printing processes, enabling the development of stretchable displays and printed circuit boards (PCB), as well as stretchable wearable electronic devices that can be attached to the human body. His academic contributions have now been recognized with this fellowship. After completing his Bachelor’s and Master’s degrees in Electrical Engineering at Seoul National University, Professor Hong received his Ph.D. from the University of Michigan. He has been actively engaged in research in the field of stretchable electronics, receiving several awards: the Scientist of the Month Award in 2019, the Merck Award in 2020, and being named in the top 100 outstanding achievements in national research and development by the Ministry of Science and ICT in 2021. In 2023, he was elected a full member of the Korean Academy of Science and Technology. The Society for Information Display has been selecting Fellows since 1963 through a rigorous evaluation process involving recommendations from past Fellows and a selection committee. Each year, fewer than 0.1% of the society's members, approximately five individuals globally, are chosen as Fellows. With the selection of Professor Hong this year, the total number of SID Fellows of Korean nationality has reached 22. The SID Fellows award ceremony will be held at the Display Week academic conference, which takes place at the McEnery Convention Center in San Jose, USA, on May 13th (local time). Original news link: https://biz.chosun.com/science-chosun/science/2024/05/08/IAGU2QKFNRGLLJI2C6AHKUG6QM/?utm_source=naver&utm_medium=original&utm_campaign=biz Source: https://ece.snu.ac.kr/ece/news?md=v&bbsidx=55554 Translated by: Jiyong Yoo, English Editor of the Department of Electrical and Computer Engineering, cyoo7@snu.ac.kr...

Research team members of this study: M.A. Dongsu Zhang of SNU ECE (1st author), Ph.D. Sangwon Im (1st author) at the front from left to right, Prof. Youngmin Kim (corresponding author), Prof. Kitae Nam (corresponding author) at the back from left to right The College of Engineering (Dean Yoosuk Hong) at Seoul National University (SNU) announced that a joint research team led by Professor Kitae Nam from the Department of Materials Science and Engineering (MSE) and Professor Youngmin Kim from the Department of Electrical and Computer Engineering (ECE) has successfully developed an AI algorithm to predict the 3D structures of nanoparticles. This achievement is a collaborative effort between the materials experiment group and the 3D AI algorithm group. AI is opening new horizons in scientific methodology and transforming the structure of industries. Following Google's protein structure prediction AI, AlphaFold, NVIDIA has introduced "Bionymo," a platform for new drug development through protein structure analysis and cellular response prediction. Innovations using AI are already happening not just in biotech but also in new materials sectors. Google DeepMind announced an AI capable of predicting new battery materials and a robotic automation laboratory that can automatically verify and develop these predicted new materials. One of the key challenges in new materials development is predicting the 3D structures of nanoparticles, which directly influence their optical, catalytic, and semiconductor properties—essentially determining the core performance of devices like sensors and batteries. Thus, it is important to predict the 3D structures and the corresponding properties of nanoparticles based on their synthesis conditions, to obtain desired characteristics of nanoparticle based devices. The newly developed AI algorithm effectively predicts the formation processes of 3D structures during the growth of nanoparticles. It successfully predicted the evolution of small seed particles to complex structures, and the division of 3D structures into chiral enantiomers. Previously, understanding the growth and morphological changes of nanoparticles at the atomic level was very challenging, but applying AI has made prediction fast and effective. The developed AI is expected to play a crucial role in the development of functional nanoparticles and devices. Particularly, this research confirmed that AI could be applied to advance the synthesis techniques of chiral gold nanoparticles, which have high potential applications in optical devices and sensors. These nanoparticles have been featured twice in Nature journal for their synthesis and device applications in 2018 and 2022, suggesting that AI can have significant potential impact on future industries. Furthermore, the research marks a significant academic contribution by applying three-dimensional cellular automata for the first time to describe the principles of nano patterns. Cellular automata, known as one of the simplest mathematical models for describing living organisms, had limitations in three-dimensional applications, but these were overcome by integration with AI. Such discoveries have the potential to provide fundamental scientific solutions for understanding not only materials, but also complex natural patterns, including the development principles of chiral enantiomers in living organisms. Professor Kitae Nam stated, "Based on the developed AI, it's now possible to implement a platform that can predict the 3D structures of nanoparticles and automatically verify them, making the AI-driven new materials development track feasible in Korea." Professor Youngmin Kim remarked, "This result demonstrates how AI can effectively understand and interpret complex phenomena at the nano level, and this methodology has great potential for expansion into broader scientific and engineering fields." The lead authors of the research, Dr. Sangwon Im and Master Dongsu Zhang, are currently working at SNU as a postdoctoral researcher and researcher, respectively. Their results have gained recognition, accepted and published in the prestigious journal in the field of material science, Nature Materials (impact factor of 41.2 in the year 2023), on May 1. This research was supported by the Future Challenge Defense Technology Development Project of the Agency for Defense Development, the International Joint Technology Development Project of the Korea Institute for Advancement of Technology, and the Scientific Challenge Interdisciplinary R&D Project of the National Research Foundation of Korea. Original Source: StartupN (https://www.startupn.kr) Source: https://ece.snu.ac.kr/ece/news?md=v&bbsidx=55543 Translated by: Jiyong Yoo, English Editor of the Department of Electrical and Computer Engineering, cyoo7@snu.ac.kr...

(From the left) Professor Jeonghun Kwak, Dr. Geunwoo Baek & Ph.D. candidate Yeounjun Kim of SNU Electrical and Computer Engineering (ECE) Department The College of Engineering (Dean Yoosuk Hong) at Seoul National University (SNU) announced on May 1st that a research team led by Professor Jeonghun Kwak, including co-first authors Dr. Geunwoo Baek and Ph.D. candidate Yeonjun Kim, has presented a resistance change mechanism for memristor* devices using core-shell structured colloidal quantum dots (QDs). The research was performed in collaboration with Professor Sooyeon Lee and Professor Sunkyu Yu from Seoul National University, and Professor Wanki Bae and Professor Jaehoon Lim from Sungkyunkwan University. Colloidal quantum dots with a core-shell structure easily allow the adjustment of electrical and optical properties depending on their size, shape, and composition, and they possess excellent luminescent properties, making them suitable for QD display manufacturing. However, due to their small size of just a few nanometers, even minor surface defects significantly affect their electrical and optical properties. The research team confirmed that controlling these surface defects on the quantum dots can modulate resistance, which they utilized in the fabrication of memristor devices. The quantum dot memristor developed by the research team demonstrated high reliability with low power operation and the capability for repetitive storage and read operations. Yet the operation mechanism of memory devices using core-shell structured quantum dots was previously unknown. To uncover the mechanism behind the change in resistance, the team introduced new methods in their device and materials research. Initially, to identify the primary charge carrier causing resistance changes within the core-shell structured quantum dots, the team inserted a layer of poly(methyl methacrylate), which is an organic insulator, either above or below the quantum dots and observed the switching characteristics. Through this approach, they determined for the first time that the primary charge carrier altering the resistance in quantum dot memristor devices is the 'electron'. Next, to determine the positions and energy levels of the defects where electrons are trapped, which are the cause of resistance changes, the research team created QD memristors using individual core-only and shell-only structured quantum dots. They measured the electrical and optical properties of these devices using various methods and compared them with actual devices. As a result, the team was the first to establish that the switching characteristics of quantum dot memristors are dictated by the trap energy levels existing on the outermost surface of the quantum dots. (Left) Comparison of QD memristors to synapse structure, (Right) Results of pattern training and inference The quantum dot memristor developed by the research team demonstrated excellent linear response characteristics required for complex processing and memory formation in synapses. Utilizing this, a simulated neural network trained on the EMNIST (Extended Modified National Institute of Standards and Technology) dataset achieved a high recognition rate of 91.46% when identifying the letters S, N, and U. Dr. Geunwoo Baek stated, "We demonstrated that high-performance memristors can be developed using core-shell structured quantum dots," and added, "By uncovering the switching mechanism, the development of quantum dot memristors is expected to become more active in the future." Professor Jeonghun Kwak noted, "It's significant that the field of research on colloidal quantum dots has expanded not only to displays but also to next-generation semiconductors like artificial neural networks. Quantum memristic memory, which can also be optically controlled, is expected to become a key technology for future photo-electric hybrid computing." Meanwhile, the results of this research were published online on May 1st in 'Nano Letters,' one of the most prestigious journals in the fields of materials, nanotechnology, and nanoscience. This research was conducted with the support of the Fundamental Research Lab program of the National Research Foundation of Korea. (From the left) Professor Jeonghun Kwak, Dr. Geunwoo Baek & Ph.D. candidate Yeounjun Kim of SNU Electrical and Computer Engineering (ECE) Department Link to paper: https://pubs.acs.org/doi/full/10.1021/acs.nanolett.4c01083 Source: https://ece.snu.ac.kr/ece/news?md=v&bbsidx=55546 Translated by: Jiyong Yoo, English Editor of the Department of Electrical and Computer Engineering, cyoo7@snu.ac.kr ...

[Reporter Mingu Gang of Edaily News] Professor Jaehyouk Choi of the ECE (Electrical and Computer Engineering) Department of Seoul National University (SNU) has been appointed as the Far East Regional Chair of the Technical Program Committee for the International Solid-State Circuits Conference (ISSCC). Since its establishment in 1954, the ISSCC has been held annually in San Francisco, USA in February, and is now celebrating its 72nd event. Renowned as the "Olympics of Semiconductor Design," it stands as the world's largest and most prestigious conference in the field of semiconductor integrated circuit design. Each year, over 4,000 semiconductor design researchers from around the world participate, representing companies, universities, and research institutes, to present the latest technologies and discuss the future of the semiconductor industry. The Technical Program Committee is responsible for paper review, selection, and program organization, and currently consists of 190 of the world's top experts in the field of semiconductor design. Technical Program Committee members are affiliated with one of the regional committees of North America, Europe, or the Far East (Asia). The Far East Regional Technical Program Committee comprises a total of 75 experts selected from Korea, Japan, Taiwan, China, Singapore, India, and other countries. Professor Jaehyouk Choi has been active as the Asia Regional Representative since 2022, following his selection as a Technical Program Subcommittee Member in 2017. Professor Choi will assume the role of '2025 ISSCC Regional Chair (ASIA TPC Chair)', scheduled to be held in San Francisco, USA, next February. In the future, he will represent Asia by attending the Executive Committee meetings, overseeing the selection of new Technical Program Committee members, deliberating on conference agendas, and overseeing regional events. Source: https://ece.snu.ac.kr/ece/news?md=v&bbsidx=55477 Translated by: Jiyong Yoo, English Editor of the Department of Electrical and Computer Engineering, cyoo7@snu.ac.kr...

The Accelerated Intelligent Systems Lab (AlSys, led by Professor Jin-Ho Lee) of the Electrical and Computer Engineering (ECE) Department has proposed a methodology for accelerating the training of Large Language Models (LLMs) using SmartSSD. Training LLMs typically requires the utilization of numerous GPUs due to the large memory requirements. However, this study suggests a technique that achieves high performance with just one GPU, by utilizing a SmartSSD that can perform computations internally. This study has been particularly recognized for its utility in situations such as fine-tuning, where training times are relatively short compared to memory requirements. As a result, it was awarded the Best Paper Award – Honorable Mention at the 2024 HPCA (International Symposium on High-Performance Computer Architecture, March 2-6, 2024, Edinburgh, Scotland), one of the top-tier conferences in the field of computer architecture. [Presented Paper] Smart-Infinity: Fast Large Language Model Training using Near-Storage Processing on a Real System, Hongsun Jang, Jaeyong Song, Jaewon Jung, Jaeyoung Park, Youngsok Kim, and Jinho Lee, HPCA, 2024 (left) HCPA 2024 award session, (right) Award certificate Source: https://ece.snu.ac.kr/ece/news?md=v&bbsidx=55465 Translated by: Jiyong Yoo, English Editor of the Department of Electrical and Computer Engineering, cyoo7@snu.ac.kr...