Kaist

Professor Minsoo Rhu Recognized as Facebook Resear..

< Professor Minsoo Rhu > Professor Minsoo Rhu from the School of Electrical Engineering was selected as the recipient of the Systems for Machine Learning Research Awards presented by Facebook. Facebook launched the award last year with the goal of funding impactful solutions in the areas of developer tookits, compilers and code generation, system architecture, memory technologies, and machine learning accelerator support. A total of 167 scholars from 100 universities representing 26 countries submitted research proposals, and Facebook selected final 10 scholars. Professor Rhu made the list with his research topic ‘A Near-Memory Processing Architecture for Training Recommendation Systems.’ He will receive 5,000 USD in research funds at the award ceremony which will take place during this year’s AI Systems Faculty Summit at the Facebook headquarters in Menlo Park, California. Professor Rhu’s submission was based on research on ‘Memory-Centric Deep Learning System Architecture’ that he carried out for three years under the auspices of Samsung Science and Technology Foundation from 2017. It was an academic-industrial cooperation research project in which leading domestic companies like Samsung Electronics and SK Hynix collaborated to make a foray into the global memory-centric smart system semiconductor market. Professor Rhu who joined KAIST in 2018 has led various systems research projects to accelerate the AI computing technology while working at NVIDIA headquarters from 2014.

New Graphene-Based Metasurface Capable of Independ..

< PhD Candidate Sangjun Han (left), Dr. Seyoon Kim (center), and Professor Min Seok Jang (right) > Researchers described a new strategy of designing metamolecules that incorporates two independently controllable subwavelength meta-atoms. This two-parametric control of the metamolecule secures the complete control of both amplitude and the phase of light. A KAIST research team in collaboration with the University of Wisconsin-Madison theoretically suggested a graphene-based active metasurface capable of independent amplitude and phase control of mid-infrared light. This research gives a new insight into modulating the mid-infrared wavefront with high resolution by solving the problem of the independent control of light amplitude and phase, which has remained a long-standing challenge. Light modulation technology is essential for developing future optical devices such as holography, high-resolution imaging, and optical communication systems. Liquid crystals and a microelectromechanical system (MEMS) have previously been utilized to modulate light. However, both methods suffer from significantly limited driving speeds and unit pixel sizes larger than the diffraction limit, which consequently prevent their integration into photonic systems. The metasurface platform is considered a strong candidate for the next generation of light modulation technology. Metasurfaces have optical properties that natural materials cannot have, and can overcome the limitations of conventional optical systems, such as forming a high-resolution image beyond the diffraction limit. In particular, the active metasurface is regarded as a technology with a wide range of applications due to its tunable optical characteristics with an electrical signal. However, the previous active metasurfaces suffered from the inevitable correlation between light amplitude control and phase control. This problem is caused by the modulation mechanism of conventional metasurfaces. Conventional metasurfaces have been designed such that a metaatom only has one resonance condition, but a single resonant design inherently lacks the degrees of freedom to independently control the amplitude and phase of light. The research team made a metaunit by combining two independently controllable metaatoms, dramatically improving the modulation range of active metasurfaces. The proposed metasurface can control the amplitude and phase of the mid-infrared light independently with a resolution beyond the diffraction limit, thus allowing complete control of the optical wavefront. The research team theoretically confirmed the performance of the proposed active metasurface and the possibility of wavefront shaping using this design method. Furthermore, they developed an analytical method that can approximate the optical properties of metasurfaces without complex electromagnetic simulations. This analytical platform proposes a more intuitive and comprehensively applicable metasurface design guideline. The proposed technology is expected to enable accurate wavefront shaping with a much higher spatial resolution than existing wavefront shaping technologies, which will be applied to active optical systems such as mid-infrared holography, high-speed beam steering devices that can be applied for LiDAR, and variable focus infrared lenses. Professor Min Seok Jang commented, "This study showed the independent control amplitude and phase of light, which has been a long-standing quest in light modulator technology. The development of optical devices using complex wavefront control is expected to become more active in the future." PhD candidate Sangjun Han and Dr. Seyoon Kim of the University of Wisconsin-Madison are the co-first authors of the research, which was published and selected as the front cover of the January 28 edition of ACS Nano titled “Complete complex amplitude modulation with electronically tunable graphene plasmonic metamolecules.” This research was funded by the Samsung Research Funding & Incubation Center for Future Technology. < Schematic image of graphene plasmonic metamolecules capable of independent amplitude and phase control of light. > < The front cover of ACS Nano published on January 28. >

New Catalyst Recycles Greenhouse Gases into Fuel a..

< Professor Cafer T. Yavuz (left), PhD Candidate Youngdong Song (center), and Researcher Sreerangappa Ramesh (right) > Scientists have taken a major step toward a circular carbon economy by developing a long-lasting, economical catalyst that recycles greenhouse gases into ingredients that can be used in fuel, hydrogen gas, and other chemicals. The results could be revolutionary in the effort to reverse global warming, according to the researchers. The study was published on February 14 in Science. “We set out to develop an effective catalyst that can convert large amounts of the greenhouse gases carbon dioxide and methane without failure,” said Cafer T. Yavuz, paper author and associate professor of chemical and biomolecular engineering and of chemistry at KAIST. The catalyst, made from inexpensive and abundant nickel, magnesium, and molybdenum, initiates and speeds up the rate of reaction that converts carbon dioxide and methane into hydrogen gas. It can work efficiently for more than a month. This conversion is called ‘dry reforming’, where harmful gases, such as carbon dioxide, are processed to produce more useful chemicals that could be refined for use in fuel, plastics, or even pharmaceuticals. It is an effective process, but it previously required rare and expensive metals such as platinum and rhodium to induce a brief and inefficient chemical reaction. Other researchers had previously proposed nickel as a more economical solution, but carbon byproducts would build up and the surface nanoparticles would bind together on the cheaper metal, fundamentally changing the composition and geometry of the catalyst and rendering it useless. “The difficulty arises from the lack of control on scores of active sites over the bulky catalysts surfaces because any refinement procedures attempted also change the nature of the catalyst itself,” Yavuz said. The researchers produced nickel-molybdenum nanoparticles under a reductive environment in the presence of a single crystalline magnesium oxide. As the ingredients were heated under reactive gas, the nanoparticles moved on the pristine crystal surface seeking anchoring points. The resulting activated catalyst sealed its own high-energy active sites and permanently fixed the location of the nanoparticles — meaning that the nickel-based catalyst will not have a carbon build up, nor will the surface particles bind to one another. “It took us almost a year to understand the underlying mechanism,” said first author Youngdong Song, a graduate student in the Department of Chemical and Biomolecular Engineering at KAIST. “Once we studied all the chemical events in detail, we were shocked.” The researchers dubbed the catalyst Nanocatalysts on Single Crystal Edges (NOSCE). The magnesium-oxide nanopowder comes from a finely structured form of magnesium oxide, where the molecules bind continuously to the edge. There are no breaks or defects in the surface, allowing for uniform and predictable reactions. “Our study solves a number of challenges the catalyst community faces,” Yavuz said. “We believe the NOSCE mechanism will improve other inefficient catalytic reactions and provide even further savings of greenhouse gas emissions.” This work was supported, in part, by the Saudi-Aramco-KAIST CO2 Management Center and the National Research Foundation of Korea. Other contributors include Ercan Ozdemir, Sreerangappa Ramesh, Aldiar Adishev, and Saravanan Subramanian, all of whom are affiliated with the Graduate School of Energy, Environment, Water and Sustainability at KAIST; Aadesh Harale, Mohammed Albuali, Bandar Abdullah Fadhel, and Aqil Jamal, all of whom are with the Research and Development Center in Saudi Arabia; and Dohyun Moon and Sun Hee Choi, both of whom are with the Pohang Accelerator Laboratory in Korea. Ozdemir is also affiliated with the Institute of Nanotechnology at the Gebze Technical University in Turkey; Fadhel and Jamal are also affiliated with the Saudi-Armco-KAIST CO2 Management Center in Korea. <Newly developed catalyst that recycles greenhouse gases into ingredients that can be used in fuel, hydrogen gas and other chemicals.> Publication: Song et al. (2020) Dry reforming of methane by stable Ni–Mo nanocatalysts on single-crystalline MgO. Science, Vol. 367, Issue 6479, pp. 777-781. Available online at http://dx.doi.org/10.1126/science.aav2412 Profile: Prof. Cafer T. Yavuz, MA, PhD yavuz＠kaist.ac.kr http://yavuz.kaist.ac.kr/ Associate Professor Oxide and Organic Nanomaterials for the Environment (ONE) Laboratory Graduate School of Energy, Environment, Water and Sustainability (EEWS) Korea Advanced Institute of Science and Technology (KAIST) http://kaist.ac.kr Daejeon, Republic of Korea Profile: Youngdong Song ydsong88＠kaist.ac.kr Ph.D. Candidate Department of Chemical and Biomolecular Engineering Korea Advanced Institute of Science and Technology (KAIST) http://kaist.ac.kr Daejeon, Republic of Korea

What Fuels a “Domino Effect” in Cancer Drug Resist..

KAIST researchers have identified mechanisms that relay prior acquired resistance to the first-line chemotherapy to the second-line targeted therapy, fueling a “domino effect” in cancer drug resistance. Their study featured in the February 7 edition of Science Advances suggests a new strategy for improving the second-line setting of cancer treatment for patients who showed resistance to anti-cancer drugs. Resistance to cancer drugs is often managed in the clinic by chemotherapy and targeted therapy. Unlike chemotherapy that works by repressing fast-proliferating cells, targeted therapy blocks a single oncogenic pathway to halt tumor growth. In many cases, targeted therapy is engaged as a maintenance therapy or employed in the second-line after front-line chemotherapy. A team of researchers led by Professor Yoosik Kim from the Department of Chemical and Biomolecular Engineering and the KAIST Institute for Health Science and Technology (KIHST) has discovered an unexpected resistance signature that occurs between chemotherapy and targeted therapy. The team further identified a set of integrated mechanisms that promotes this kind of sequential therapy resistance. “There have been multiple clinical accounts reflecting that targeted therapies tend to be least successful in patients who have exhausted all standard treatments,” said the first author of the paper Mark Borris D. Aldonza. He continued, “These accounts ignited our hypothesis that failed responses to some chemotherapies might speed up the evolution of resistance to other drugs, particularly those with specific targets.” Aldonza and his colleagues extracted large amounts of drug-resistance information from the open-source database the Genomics of Drug Sensitivity in Cancer (GDSC), which contains thousands of drug response data entries from various human cancer cell lines. Their big data analysis revealed that cancer cell lines resistant to chemotherapies classified as anti-mitotic drugs (AMDs), toxins that inhibit overacting cell division, are also resistant to a class of targeted therapies called epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs). In all of the cancer types analyzed, more than 84 percent of those resistant to AMDs, representatively ‘paclitaxel’, were also resistant to at least nine EGFR-TKIs. In lung, pancreatic, and breast cancers where paclitaxel is often used as a first-line, standard-of-care regimen, greater than 92 percent showed resistance to EGFR-TKIs. Professor Kim said, “It is surprising to see that such collateral resistance can occur specifically between two chemically different classes of drugs.” To figure out how failed responses to paclitaxel leads to resistance to EGFR-TKIs, the team validated co-resistance signatures that they found in the database by generating and analyzing a subset of slow-doubling, paclitaxel-resistant cancer models called ‘persisters’. The results demonstrated that paclitaxel-resistant cancers remodel their stress response by first becoming more stem cell-like, evolving the ability to self-renew to adapt to more stressful conditions like drug exposures. More surprisingly, when the researchers characterized the metabolic state of the cells, EGFR-TKI persisters derived from paclitaxel-resistant cancer cells showed high dependencies to energy-producing processes such as glycolysis and glutaminolysis. “We found that, without an energy stimulus like glucose, these cells transform to becoming more senescent, a characteristic of cells that have arrested cell division. However, this senescence is controlled by stem cell factors, which the paclitaxel-resistant cancers use to escape from this arrested state given a favorable condition to re-grow,” said Aldonza. Professor Kim explained, “Before this research, there was no reason to expect that acquiring the cancer stem cell phenotype that dramatically leads to a cascade of changes in cellular states affecting metabolism and cell death is linked with drug-specific sequential resistance between two classes of therapies.” He added, “The expansion of our work to other working models of drug resistance in a much more clinically-relevant setting, perhaps in clinical trials, will take on increasing importance, as sequential treatment strategies will continue to be adapted to various forms of anti-cancer therapy regimens.” This study was supported by the Basic Science Research Program of the National Research Foundation of Korea (NRF-2016R1C1B2009886), and the KAIST Future Systems Healthcare Project (KAISTHEALTHCARE42) funded by the Korean Ministry of Science and ICT (MSIT). Undergraduate student Aldonza participated in this research project and presented the findings as the lead author as part of the Undergraduate Research Participation (URP) Program at KAIST. < Figure 1. Schematic overview of the study. > < Figure 2. Big data analysis revealing co-resistance signatures between classes of anti-cancer drugs. > Publication: Aldonza et al. (2020) Prior acquired resistance to paclitaxel relays diverse EGFR-targeted therapy persistence mechanisms. Science Advances, Vol. 6, No. 6, eaav7416. Available online at http://dx.doi.org/10.1126/sciadv.aav7416 Profile: Prof. Yoosik Kim, MA, PhD ysyoosik＠kaist.ac.kr https://qcbio.kaist.ac.kr/ Assistant Professor Bio Network Analysis Laboratory Department of Chemical and Biomolecular Engineering Korea Advanced Institute of Science and Technology (KAIST) http://kaist.ac.kr Daejeon, Republic of Korea Profile: Mark Borris D. Aldonza borris＠kaist.ac.kr Undergraduate Student Department of Biological Sciences Korea Advanced Institute of Science and Technology (KAIST) http://kaist.ac.kr Daejeon, Republic of Korea

Professor Youngseok Ju Awarded the 13th ASAN Award..

<Professor Youngseok Ju> Professor Youngseok Ju from the Graduate School of Medical Science and Engineering was selected for the 13th ASAN Award for Young Medical Scientists under the age of 40. Professor Ju will receive 50 million won in prize money. The ASAN Foundation established this Award in 2007 to encourage young medical scientists who accomplished outstanding achievements in basic and clinical medicine. The winners are chosen based on a comprehensive assessment of consistency and originality, domestic and international impact, and contributions to medical development and fostering future generations. Professor Ju is known for having identified the generation principle of cancer genome mutations. In particular, he is recognized for his contributions to the development of cancer prevention, diagnosis, and treatment, by having proven that some cases of lung cancer can occur from destructive changes in chromosomes in lung cells regardless of smoking. The award ceremony will be held on March 19 in Seoul. The other award will be given to Professor Yong-Ho Lee from the Yonsei University College of Medicine.

Blood-Based Multiplexed Diagnostic Sensor Helps to..

A research team at KAIST reported clinically accurate multiplexed electrical biosensor for detecting Alzheimer’s disease by measuring its core biomarkers using densely aligned carbon nanotubes. Alzheimer’s disease is the most prevalent neurodegenerative disorder, affecting one in ten aged over 65 years. Early diagnosis can reduce the risk of suffering the disease by one-third, according to recent reports. However, its early diagnosis remains challenging due to the low accuracy but high cost of diagnosis. Research team led by Professors Chan Beum Park and Steve Park described an ultrasensitive detection of multiple Alzheimer's disease core biomarker in human plasma. The team have designed the sensor array by employing a densely aligned single-walled carbon nanotube thin films as a transducer. The representative biomarkers of Alzheimer's disease are beta-amyloid42, beta-amyloid40, total tau protein, phosphorylated tau protein and the concentrations of these biomarkers in human plasma are directly correlated with the pathology of Alzheimer’s disease. The research team developed a highly sensitive resistive biosensor based on densely aligned carbon nanotubes fabricated by Langmuir-Blodgett method with a low manufacturing cost. Aligned carbon nanotubes with high density minimizes the tube-to-tube junction resistance compared with randomly distributed carbon nanotubes, which leads to the improvement of sensor sensitivity. To be more specific, this resistive sensor with densely aligned carbon nanotubes exhibits a sensitivity over 100 times higher than that of conventional carbon nanotube-based biosensors. By measuring the concentrations of four Alzheimer’s disease biomarkers simultaneously Alzheimer patients can be discriminated from health controls with an average sensitivity of 90.0％, a selectivity of 90.0％ and an average accuracy of 88.6％. This work, titled “Clinically accurate diagnosis of Alzheimer’s disease via multiplexed sensing of core biomarkers in human plasma”, were published in Nature Communications on January 8th 2020. The authors include PhD candidate Kayoung Kim and MS candidate Min-Ji Kim. Professor Steve Park said, “This study was conducted on patients who are already confirmed with Alzheimer’s Disease. For further use in practical setting, it is necessary to test the patients with mild cognitive impairment.” He also emphasized that, “It is essential to establish a nationwide infrastructure, such as mild cognitive impairment cohort study and a dementia cohort study. This would enable the establishment of world-wide research network, and will help various private and public institutions.” This research was supported by the Ministry of Science and ICT, Human Resource Bank of Chungnam National University Hospital and Chungbuk National University Hospital. (Figure: A schematic diagram of a high-density aligned carbon nanotube-based resistive sensor that distinguishes patients with Alzheimer’s Disease by measuring the concentration of four biomarkers in the blood.) -Profile Professor Steve Park stevepark＠kaist.ac.kr Department of Materials Science and Engineering http://steveparklab.kaist.ac.kr/ KAIST Professor Chan Beum Park parkcb at kaist.ac.kr Department of Materials Science and Engineering http://biomaterials.kaist.ac.kr/ KAIST

Cancer Cell Reversion May Offer a New Approach to ..

A novel approach to reverse the progression of healthy cells to malignant ones may offer a more effective way to eradicate colorectal cancer cells with far fewer side effects, according to a team of researchers based in South Korea. Colorectal cancer, or cancer of the colon, is the third most common cancer in men and the second most common in women worldwide. South Korea has the second highest incident rate of colorectal cancer in the world, topped only by Hungary, according to the World Cancer Research Fund. Their results were published as a featured cover article on January 2 in Molecular Cancer Research, a journal of the American Association for Cancer Research. Led by Kwang-Hyun Cho, a professor and associate vice president of research at KAIST , the researchers used a computational framework to analyze healthy colon cells and colorectal cancer cells. They found that some master regulator proteins involved in cellular replication helped healthy colon cells mature, or differentiate into their specific cell type, and remain healthy. One particular protein, called SETDB1, suppressed the helpful proteins, forcing new cells to remain in a state of immaturity with the potential to become cancerous. “This suggests that differentiated cells have an inherent resistance mechanism against malignant transformation and indicates that cellular reprogramming is indispensable for malignancy,” said Cho. “We speculated that malignant properties might be eradicated if the tissue-specific gene expression is reinstated — if we repress SETDB1 and allow the colon cells to mature and differentiate as they would normally.”  1. Image caption: Expression of SETDB1 in colorectal cancer tissues and cellular differentiation of patient-derived colon cancer organoids upon SETDB1 depletion. 2. Image credit: Kwang-Hyun Cho, KAIST 3. Image restriction: News organizations may use or redistribute this image, with proper attribution, as part of news coverage of this paper only. Using human-derived cells, Cho and his team targeted the tissue-specific gene expression programs identified in their computational analysis. These are the blueprints for the proteins that eventually help immature cells differentiate into tissue-specific cell types, such as colon cells. When a person has a genetic mutation, or has exposure to certain environmental factors, this process can go awry, leading to an overexpression of unhelpful proteins, such as SEDTB1. The researchers specifically reduced the amount of SEDTB1 in these tissue-specific gene expression programs, which allowed the cells to mature and fully differentiate into colon cells. “Our experiment also shows that SETDB1 depletion combined with cytotoxic drugs might be potentially beneficial to anticancer treatment,” Cho said. Cytotoxic drugs are often used for cancer treatment because the type of medicine contains chemicals that are toxic to cancer cells which can prevent them from replicating or growing. He noted that this combination could be more effective in treating cancer by transforming the cancer cell state into a less malignant or resistant state. He eventually pursues a cancer reversion therapy alone instead of conventional cytotoxic drug therapy since the cancer reversion therapy can provide a much less painful experience for patients with cancer who often have severe side effects from treatments intended to kill off cancerous cells, such as chemotherapy. The researchers plan to continue studying how to return cancer cells to healthier states, with the ultimate goal of translating their work to therapeutic treatment for patients with colorectal cancer. “I think our study of cancer reversion would eventually change the current medical practice of treating cancer toward the direction of keeping the patient’s quality of life while minimizing the side effects of current anti-cancer therapies,” Cho said. ### This work was funded by KAIST and the National Research Foundation of Korea grants funded by the Korean government, the Ministry of Science and Information and Communication Technology. Other authors include Soobeom Lee, Chae Young Hwang and Dongsan Kim, all of whom are affiliated with the Laboratory for Systems Biology and Bio-Inspired Engineering in the Department of Bio and Brain Engineering at KAIST; Chansu Lee and Sung Noh Hong, both with the Department of Medicine, and Seok-Hyung Kim of the Department of Pathology in the Samsung Medical Center at the Sungkyunkwan University School of Medicine. -Profile Professor Kwang-Hyun Cho ckh＠kaist.ac.kr http://sbie.kaist.ac.kr/ Department of Bio and Brain Engineering KAIST https://www.kaist.ac.kr

New Insights into How the Human Brain Solves Compl..

A new study on meta reinforcement learning algorithms helps us understand how the human brain learns to adapt to complexity and uncertainty when learning and making decisions. A research team, led by Professor Sang Wan Lee at KAIST jointly with John O’Doherty at Caltech, succeeded in discovering both a computational and neural mechanism for human meta reinforcement learning, opening up the possibility of porting key elements of human intelligence into artificial intelligence algorithms. This study provides a glimpse into how it might ultimately use computational models to reverse engineer human reinforcement learning. This work was published on Dec 16, 2019 in the journal Nature Communications. The title of the paper is “Task complexity interacts with state-space uncertainty in the arbitration between model-based and model-free learning.” Human reinforcement learning is an inherently complex and dynamic process, involving goal setting, strategy choice, action selection, strategy modification, cognitive resource allocation etc. This a very challenging problem for humans to solve owing to the rapidly changing and multifaced environment in which humans have to operate. To make matters worse, humans often need to often rapidly make important decisions even before getting the opportunity to collect a lot of information, unlike the case when using deep learning methods to model learning and decision-making in artificial intelligence applications. In order to solve this problem, the research team used a technique called 'reinforcement learning theory-based experiment design' to optimize the three variables of the two-stage Markov decision task - goal, task complexity, and task uncertainty. This experimental design technique allowed the team not only to control confounding factors, but also to create a situation similar to that which occurs in actual human problem solving. Secondly, the team used a technique called ‘model-based neuroimaging analysis.’ Based on the acquired behavior and fMRI data, more than 100 different types of meta reinforcement learning algorithms were pitted against each other to find a computational model that can explain both behavioral and neural data. Thirdly, for the sake of a more rigorous verification, the team applied an analytical method called ‘parameter recovery analysis,’ which involves high-precision behavioral profiling of both human subjects and computational models. In this way, the team was able to accurately identify a computational model of meta reinforcement learning, ensuring not only that the model’s apparent behavior is similar to that of humans, but also that the model solves the problem in the same way as humans do. The team found that people tended to increase planning-based reinforcement learning (called model-based control), in response to increasing task complexity. However, they resorted to a simpler, more resource efficient strategy called model-free control, when both uncertainty and task complexity were high. This suggests that both the task uncertainty and the task complexity interact during the meta control of reinforcement learning. Computational fMRI analyses revealed that task complexity interacts with neural representations of the reliability of the learning strategies in the inferior prefrontal cortex. These findings significantly advance understanding of the nature of the computations being implemented in the inferior prefrontal cortex during meta reinforcement learning as well as providing insight into the more general question of how the brain resolves uncertainty and complexity in a dynamically changing environment. Identifying the key computational variables that drive prefrontal meta reinforcement learning, can also inform understanding of how this process might be vulnerable to break down in certain psychiatric disorders such as depression and OCD. Furthermore, gaining a computational understanding of how this process can sometimes lead to increased model-free control, can provide insights into how under some situations task performance might break down under conditions of high cognitive load. Professor Lee said, “This study will be of enormous interest to researchers in both the artificial intelligence and human/computer interaction fields since this holds significant potential for applying core insights gleaned into how human intelligence works with AI algorithms.” This work was funded by the National Institute on Drug Abuse, the National Research Foundation of Korea, the Ministry of Science and ICT, Samsung Research Funding Center of Samsung Electronics. Figure 1 (modified from the figures of the original paper doi:10.1038/s41467-019-13632-1). Computations implemented in the inferior prefrontal cortex during meta reinforcement learning. (A) Computational model of human prefrontal meta reinforcement learning (left) and the brain areas whose neural activity patterns are explained by the latent variables of the model. (B) Examples of behavioral profiles. Shown on the left is choice bias for different goal types and on the right is choice optimality for task complexity and uncertainty. (C) Parameter recoverability analysis. Compared are the effect of task uncertainty (left) and task complexity (right) on choice optimality. -Profile Professor Sang Wan Lee sangwan＠kaist.ac.kr Department of Bio and Brain Engineering Director, KAIST Center for Neuroscience-inspired AI KAIST Institute for Artificial Intelligence (http://aibrain.kaist.ac.kr) KAIST Institute for Health, Science, and Technology KAIST (https://www.kaist.ac.kr)

Professor Junil Choi Receives Stephen O. Rice Priz..

< Professor Junil Choi (second from the left) > Professor Junil Choi from the School of Electrical Engineering received the Stephen O. Rice Prize at the Global Communications Conference (GLOBECOM) hosted by the Institute of Electrical and Electronics Engineers (IEEE) in Hawaii on December 10, 2019. The Stephen O. Rice Prize is awarded to only one paper of exceptional merit every year. The IEEE Communications Society evaluates all papers published in the IEEE Transactions on Communications journal within the last three years, and marks each paper by aggregating its scores on originality, the number of citations, impact, and peer evaluation. Professor Choi won the prize for his research on one-bit analog-to-digital converters (ADCs) for multiuser massive multiple-input and multiple-output (MIMO) antenna systems published in 2016. In his paper, Professor Choi proposed a technology that can drastically reduce the power consumption of the multiuser massive MIMO antenna systems, which are the core technology for 5G and future wireless communication. Professor Choi’s paper has been cited more than 230 times in various academic journals and conference papers since its publication, and multiple follow-up studies are actively ongoing. In 2015, Professor Choi received the IEEE Signal Processing Society Best Paper Award, an award equals to the Stephen O. Rice Prize. He was also selected as the winner of the 15th Haedong Young Engineering Researcher Award presented by the Korean Institute of Communications and Information Sciences (KICS) on December 6, 2019 for his outstanding academic achievements, including 34 international journal publications and 26 US patent registrations.

Professor Shin-Hyun Kim Receives the Young Scienti..

Professor Shin-Hyun Kim from the Department of Chemical and Biomolecular Engineering received the Young Scientist Award from the Korean Academy of Science and Technology. The Young Scientist Award is presented to a promising young Korean scientist under the age of 40 who shows significant potential, passion, and remarkable achievement. Professor Kim was lauded for his research of intelligent soft materials. By applying his research, he developed a capsule sensor material that can not only be used for sensors, but also for displays, color aesthetics, anti-counterfeit technology, residual drug detection, and more. The award ceremony took place on December 14 at the Gwacheon National Science Museum. The Korean minister of Science and ICT delivered words of encouragement, reminding everyone that “the driving force behind creative performance of scientists is the provision of continuous support.” He added, “Researchers of Korea deserve greater public attention and support.”

Team Geumo Wins Consecutive Victories in K-Cyber S..

< Professor Sang Kil Cha > < Masters Candidate Kangsu Kim and Researcher Corentin Soulet > Team Geumo, led by Professor Sang Kil Cha from the Graduate School of Information Security, won the K-Cyber Security Challenge in the AI-based automatic vulnerability detection division for two consecutive years in 2018 and 2019. The K-Cyber Security Challenge is an inter-machine hacking competition. Participants develop and operate AI-based systems that are capable of independently identifying software vulnerabilities and gaining operating rights through hacking. The K-Cyber Security Challenge, inspired by the US Cyber Grand Challenge launched by the Defense Advanced Research Projects Agency (DARPA), is hosted by the Ministry of Science and ICT and organized by the Korea Internet and Security Agency. Researcher Corentin Soulet of the School of Computing and master’s student Kangsu Kim of the Graduate School of Information Security teamed up for the competition. Professor Cha, who has led the research on software and systems security since his days at Carnegie Mellon University, succeeded in establishing a world-class system using domestic technology. In a recent collaboration with the Cyber Security Research Center, Professor Cha achieved a ten-fold increase in the speed of binary analysis engines, a key component of AI-based hacking systems. For this accomplishment, he received the Best Paper Award at the 2019 Network and Distributed System Security Workshop on Binary Analysis Research (NDSS BAR). Kangsu Kim said, "It is a great honor to win the competition two years in a row. I will continue to work hard and apply my knowledge to serve society.”

Two Professors Receive Awards from the Korea Robot..

< Professor Jee-Hwan Ryu and Professor Ayoung Kim > The Korea Robotics Society (KROS), in recognition of their achievements and contributions to the development of robotic technology and industry in 2019, awarded two KAIST professors from the Department of Civil and Environmental Engineering. Professor Jee-Hwan Ryu has been actively engaged in researching the field of teleoperation, and this led him to win the KROS Robotics Innovation (KRI) Award. The KRI Award was newly established in 2019 by the KROS, in order to encourage researchers who have made innovative achievements in robotics. Professor Ryu shared the honor of being the first winners of this award with Professor Jaeheung Park of Seoul National University. Professor Ayoung Kim, from the same department, received the Young Investigator Award presented to emerging robitics researchers under 40 years of age.