Archives: 연구 > 연구성과

EE Professor Kim Joo-Young Developed A ChatGPT Core AI Semiconductor with A 2.4-fold Improvement in Price efficiency

 

 

 

The ChatGPT released by OpenAI has captured global attention, and everyone is closely observing the changes this technology will bring out.

This technology is based on large language models (LLM), which represent an unprecedented scale of artificial intelligence (AI) models compared to conventional AI.

However, the operation of these models requires a significant number of high-performance GPUs, leading to astronomical computing costs.  

 

KAIST (President: Lee Kwang-Hyung) announced that research team led by EE Professor Kim Joo-Young Kim has successfully developed an AI semiconductor that efficiently accelerates the inference operations of large language models, which play a crucial role in ChatGPT. 

The developed AI semiconductor, named the ‘Latency Processing Unit (LPU),’ efficiently accelerates the inference operations of large language models. It incorporates a high-speed computing engine capable of maximizing memory bandwidth utilization and performing all necessary inference computations rapidly.

Additionally, it comes equipped built-in networking capabilities, making it easily expandable with multiple accelerators. This LPU-based acceleration appliance server achieved up to a 50% higher performance and approximately 2.4 times better performance-to-price ratio compared to a supercomputer based on the industry-leading high-performance GPU, NVIDIA A100.

 

This advancement holds the potential to replace high-performance GPUs in data centers that are experiencing a rapid surge in demand for generative AI services. This research was conducted by Professor HyperExcel Co., founded by Professor Kim Joo-Young and achieved the remarkable accomplishment of receiving the “Engineering Best Presentation Award” at the International Design Automation Conference (DAC 2023) held in San Francisco on July 12th (U.S. time).

DAC is a prestigious international conference in the field of semiconductor design, particularly showcasing global semiconductor design technologies related to Electronic Design Automation (EDA) and Semiconductor Intellectual Property (IP).

DAC attracts participation from renowned semiconductor design companies such as Intel, NVIDIA, AMD, Google, Microsoft, Samsung, TSMC, as well as top universities including Harvard, MIT, and Stanford.

 

Among the world’s notable semiconductor technologies, Professor Kim’s team stands out as the sole recipient of an award for AI semiconductor technology tailored for large language models.

This award acknowledges their AI semiconductor solution as a groundbreaking means to drastically reduce the substantial costs associated with inference operations for large language models on the global stage.

 

Professor Kim stated, “With the new processor ‘LPU’ for future large AI computations, I intend to pioneer the global market and take a lead over big tech companies in terms of technological prowess.”

(Note: The provided translation is an elaboration and summary of the original text for clarity and readability.)

 

Professor Joo-Young Kim’s Research Team Published Article in CACM Magazine: “South Korea’s Nationwide Effort for AI Semiconductor Industry”

 

Recently, the research team led by Professor Joo-Young Kim published an article titled “South Korea’s Nationwide Effort for AI Semiconductor Industry” in CACM (Communications of the ACM), one of the leading monthly academic journals in the field of computer science.

 

 

 

Professor Shim Hyun-Chul(Principal Investigators), Professor Kim Min-Jun, and Collaborative Research Team Develop World’s First Humanoid Pilot, “PiBot.”

 

Recently, there has been a lot of attention on KAIST’s development of the humanoid pilot “PiBot” following the legendary pilot Maverick’s flights in the recent movie “Top Gun: Maverick.”

KAIST (President Lee Kwang-Hyung) announced on the 19th that they plan to develop a practical humanoid robot capable of directly piloting an aircraft based on understanding technical manuals written in natural language.

 

The collaborative research team, led by KAIST Professor Shim Hyun-Chul and including Professor Choo Jae-Gul, Professor Yoon Kuk-Jin, and Professor Kim Minjun, applied artificial intelligence and robotics technology to develop an “NLP-based humanoid pilot robot,” as part of the Future Challenge Project, which involves reading and comprehending pilot manuals written in general language.

 

The robot was able to sit in the cockpit of a conventional aircraft without any modification and directly operate various controls, showcasing a different approach from conventional aircraft’s autopilot systems or unmanned aerial vehicles that can only perform unmanned flights.

 

The pilot robot being developed by the research team can memorize and flawlessly operate the entire global Jeppesen Chart, which is impossible for human pilots.

It can also immediately respond by leveraging the recent breakthroughs in ChatGPT technology, recalling aircraft operation manuals and Quick Reference Handbook (QRH) procedures, and calculating real-time safe routes based on the aircraft’s flight status.

 

The robot can accurately perceive the internal and external conditions of the cockpit and the aircraft using the onboard cameras, precisely manipulate various switches, and control its robotic arms and hands with high precision even in highly vibrating aircraft environments.

 

Currently, the pilot robot is capable of performing all aircraft operations, including starting the engine, taxiing, takeoff, cruising, and landing, in a flight simulator. The research team plans to apply the pilot robot to an actual light aircraft for validation.

 

Professor Shim Hyun-Chul, the project leader, stated, “The humanoid pilot robot has high practicality and applicability, as it can immediately perform autonomous flights without any modification to existing aircraft. It can also operate a variety of devices, including aircraft, cars, and armored vehicles, making it highly useful in situations where there is a shortage of human resources.”

 

The ongoing project is supported by the Future Challenge Project from the Defense Science Institute (total funding of 5.7 billion KRW) and has been in development since 2022 through the collaboration of Professor Shim Hyun-Chul from the Department of Electrical and Electronic Engineering at KAIST (project leader), Professor Choo Jae-Gul from the AI Graduate School, Professor Yoon Kuk-Jin from the Department of Mechanical Engineering, and Professor Kim Min-Jun from the Department of Electrical and Electronic Engineering.

 

The project is expected to be completed in 2026 and explore commercialization plans for civilian and military applications.

 

 

Professor Insu Yun has received the Frontiers of Science Award at the ICBS (International Conference on Basic Science).

 

[Professor Insu Yun]

 

 

OLEDs are light-emitting devices that are being utilized as cutting-edge display technology, with various advantages such as clear image quality, thin profile, and the capability to fabricate flexible devices such as foldable or rollable displays.

In such applications, securing sufficient efficiency and lifespan of the red, green, and blue light sources that make up the three primary colors of light while also securing high color purity of the three primary colors is critical, but the technology to secure all these requirements in blue OLED devices is considered to be a big challenge.

Professor Seunghyup Yoo’s research team from the School of EE in collaboration with Professor Kim Yoon-hee’s research team from the Department of Chemistry at Gyeongsang National University succeeded in implementing a deep-blue organic light-emitting diode (OLED) device with the world’s highest level of efficiency.

Dr. Hyungseok Kim from Professor Yoo’s lab at KAIST (currently a visiting researcher from Kyushu University), Dr. Hyeong-jin Cheon from Gyeongsang National University (currently a visiting researcher at Imperial College London), and Donggyun Lee, a Ph.D. candidate from KAIST (in Professor Yoo’s lab), were the joint first authors of this study.

The research was published in the online version of the international journal ‘Science Advances’ on May 31, 2023. (Paper title: Toward highly efficient deep-blue OLEDs: Tailoring the multiresonance-induced TADF molecules for suppressed excimer formation and near-unity horizontal dipole ratio).

The team focused on the design of boron-based thermally activated delayed fluorescence materials, which are emerging as the next-generation emitters for the implementation of deep-blue OLED devices. These materials have the advantage of achieving excellent color purity, but due to their flat molecular structure, strong interactions between molecules can limit deep blue colors to low concentrations.

When the concentration of luminescent molecules is increased to secure sufficient efficiency for OLED devices, the issue of not being able to fully exploit the color purity advantage inherent in the emitters arises.

The research team not only simplified the synthesis process and improved the low yield compared to the conventional boron-based materials known to be very difficult to synthesize, but also successfully unveiled a molecular structure that can suppress intermolecular interactions from the perspective of molecular dynamics.

By implementing this through molecular design, the team solved the challenging issue of color purity and efficiency degradation. Unlike previous trials and errors that were carried out empirically, the research team established a comprehensive and analytical methodology, theoretically predicting structures that can lead to maximum efficiency, and successfully implemented a deep-blue OLED with more than 35% maximum external quantum efficiency by grafting the high-efficiency organic luminescent material they designed onto the device structure.

This is one of the world’s highest levels of efficiency for deep-blue OLED unit devices at this wavelength.

Professor Yoo commented, “Securing high-efficiency deep-blue OLED technology is one of the essential tasks in perfecting OLED displays as the ultimate technology, and this study is a good example that demonstrates the importance of systematic integrated research and collaboration between materials and device groups in solving challenges.”

 

 

 

On June 26, 2023, the Ministry of Science and ICT (MSIT) presented the “Minister’s Award” to professor Youngik Sohn during the opening ceremony of the ‘Quantum Korea 2023 Conference’ held at Dongdaemun Design Plaza (DDP) in Seoul.

Professor Sohn was recognized for his contributions to advancing the Artificial Intelligence (AI) semiconductor industry.

[Awards Ceremonies]

Development of transparent flexible OLED from school of EE Kyung Cheol Choi’s laboratory

 

 

Ph.D. Candidate Sungjun Ahn(advisor Joonhyuk Kang) won Best Student Paper Award at IEEE BMSB 2023 Conference

 

Ahn Sungjun, a Ph.D. candidate in EE Professor Kang Joonhyuk’s lab, received the Best Student Paper at the IEEE BMSB 2023 Conference in Beijing, China.

The IEEE BMSB (IEEE International Symposium on Broadband Multimedia Systems and Broadcasting) is the largest academic conference in the world specializing in broadcasting technology, organized by the Broadcast Technology Society of IEEE.

 

(from left: Ahn Sungjun, Ph.D. candidae, Professor Kang Joonhyuk)

 

Ahn Sungjun (Paper title: Adaptive Modification of Fading Channel Models for Terrestrial SFN Environments), provides essential channel models for performance analysis of Single Frequency Networks (SFN), which is widely adopted in second-generation terrestrial broadcasting systems.

This study presents a technique that adapts measurement-based channel models to apply them to the receiving environment. This research is expected to have broad applications in the worldwide standardization, technological development, and performance analysis of terrestrial broadcasting and communication systems.

Ahn Sungjun expressed in his acceptance speech, “I will continue to strive for excellent research outcomes through active international collaboration.” We express our gratitude and congratulations to the professor, the awardee, and the research lab.

 

 

 

 

 

 

Songhyeon Kuk, Ph.D student from professor Sanghyeon Kim’s lab, won the Best Student Paper at the 2023 International Memory Workshop (IMW). 

IMW is an international conference dealing with technology development for memory technology.

It is a competitive conference with an acceptance rate of around 30% (29% this year) and was held in Monterey, USA this year.

PhD Candidate Songhyeon Kuk’s paper proposed the idea of using P-channel to reduce the performance degradation of the conventional ferroelectric field effect transistors (FET).

It was suggested that the P-Channel ferroelectric FET could be a candidate for next-generation NAND flash memory with excellent performance, and in recognition of this value, Songhyeon Kuk won the Best student paper award.

In addition, Songhyun Kuk presented a paper on the method of designing ferroelectric FET using next-generation semiconductor processes at the VLSI symposium on Technology and Circuits held last week.

 

[Mr. Gangmin Cho, Dr. Yonghwi Kwon, Dr. Pervaiz Kareem, and Prof. Youngsoo Shin from left]

 

The IEEE Transactions on Semiconductor Manufacturing publishes the latest advances related to the manufacture of microelectronic and photonic components and integrated systems, including photovoltaic devices and micro-electro-mechanical systems.

Among around 100 papers published in 2022 at IEEE TSM, their paper has been selected as one of the three papers for Best Paper Award – Honorable Mention.

Dr. Yonghwi Kwon and Prof. Youngsoo Shin, in particular, were recipients of 2021 BPA from the same journal, i.e. they received BPA in 2021 and BPA – Honorable Mention in 2022, a rare and extraordinary achievement.

 

– Title: 2022 IEEE TSM Best Paper Award – Honorable Mention 

– Paper Title: Integrated Test Pattern Extraction and Generation for Accurate Lithography Modeling

– Authors: Gangmin Cho, Yonghwi Kwon, Pervaiz Kareem, and Youngsoo Shin (Advisor professor)

– Journal: IEEE Transactions on Semiconductor Manufacturing