BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Biomedical Mathematics Group - ECPv6.17.0//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://www.ibs.re.kr/bimag
X-WR-CALDESC:Events for Biomedical Mathematics Group
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Seoul
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:KST
DTSTART:20250101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Seoul:20260605T100000
DTEND;TZID=Asia/Seoul:20260605T120000
DTSTAMP:20260527T003414Z
CREATED:20260429T070216Z
LAST-MODIFIED:20260527T003414Z
UID:12396-1780653600-1780660800@www.ibs.re.kr
SUMMARY:Chaos Meets Attention: Transformers for Large-Scale Dynamical Prediction- Gyuyoung Hwang
DESCRIPTION:In this talk\, we discuss the paper “Chaos Meets Attention: Transformers for Large-Scale Dynamical Prediction” by Yi He et al.\, ICML Poster\, 2025. \nAbstract: \nGenerating long-term trajectories of dissipative chaotic systems autoregressively is a highly challenging task. The inherent positive Lyapunov exponents amplify prediction errors over time. Many chaotic systems possess a crucial property — ergodicity on their attractors\, which makes long-term prediction possible. State-of-the-art methods address ergodicity by preserving statistical properties using optimal transport techniques. However\, these methods face scalability challenges due to the curse of dimensionality when matching distributions. To overcome this bottleneck\, we propose a scalable transformer-based framework capable of stably generating long-term high-dimensional and high-resolution chaotic dynamics while preserving ergodicity. Our method is grounded in a physical perspective\, revisiting the Von Neumann mean ergodic theorem to ensure the preservation of long-term statistics in the L2 space. We introduce novel modifications to the attention mechanism\, making the transformer architecture well-suited for learning large-scale chaotic systems. Compared to operator-based and transformer-based methods\, our model achieves better performances across five metrics\, from short-term prediction accuracy to long-term statistics. In addition to our methodological contributions\, we introduce new chaotic system benchmarks: a machine learning dataset of 140 snapshots of turbulent channel flow and a processed high-dimensional Kolmogorov Flow dataset\, along with various evaluation metrics for both short- and long-term performances. Both are well-suited for machine learning research on chaotic systems.
URL:https://www.ibs.re.kr/bimag/event/bridging-known-and-unknown-dynamics-by-transformer-based-machine-learning-inference-from-sparse-observations-gyuyoung-hwang/
LOCATION:B232 Seminar Room\, IBS\, 55 Expo-ro Yuseong-gu\, Daejeon\, Daejeon\, 34126\, Korea\, Republic of
CATEGORIES:Journal Club
ORGANIZER;CN="Jae Kyoung Kim":MAILTO:jaekkim@kaist.ac.kr
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Seoul:20260619T100000
DTEND;TZID=Asia/Seoul:20260619T120000
DTSTAMP:20260614T040037Z
CREATED:20260520T075146Z
LAST-MODIFIED:20260614T040037Z
UID:12428-1781863200-1781870400@www.ibs.re.kr
SUMMARY:Inferring circadian phases and quantifying biological desynchrony across single-cell transcriptomes - Dongju Lim
DESCRIPTION:In this talk\, we discuss the paper “Inferring circadian phases and quantifying biological desynchrony across single-cell transcriptomes” by Andrea Salati et al.\, bioRxiv\, 2026. \n  \nAbstract: \nSingle-cell RNA sequencing (scRNA-seq) reveals heterogeneity in circadian clock states across individual cells\, yet accurately inferring circadian phase and distinguishing biological desynchrony from technical noise remains challenging. Here\, we introduce scRitmo\, a probabilistic framework that infers single-cell circadian phases from mRNA count data\, providing both a point estimate and a posterior uncertainty for each cell. A simulationcalibrated variance decomposition separates the observed phase dispersion into biological and technical components\, enabling direct estimation of intercellular desynchrony. We validate scRitmo using deeply sequenced unsynchronized fibroblasts\, where inferred transcriptomic phases accurately predict protein-level oscillations of a circadian reporter. Applied to murine scRNA-seq datasets from liver\, aorta\, and skin\, scRitmo outperforms existing methods and reveals cell-type-specific levels of phase coherence. In SABER-FISH time-series data\, the method recovers the progressive accumulation of desynchrony following synchronization\, and in Drosophila clock neurons it captures cell-type-specific phase shifts and the expected increase in phase dispersion under constant darkness relative to light-dark entrainment. Together\, scRitmo provides a principled approach for quantifying circadian (de)synchrony from transcriptomic data\, decoupling biological phase variability from measurement noise across tissues\, organisms\, and experimental conditions.
URL:https://www.ibs.re.kr/bimag/event/insulin-resistance-prediction-from-wearables-and-routine-blood-biomarkers-dongju-lim/
LOCATION:B232 Seminar Room\, IBS\, 55 Expo-ro Yuseong-gu\, Daejeon\, Daejeon\, 34126\, Korea\, Republic of
CATEGORIES:Journal Club
ORGANIZER;CN="Jae Kyoung Kim":MAILTO:jaekkim@kaist.ac.kr
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Seoul:20260626T093000
DTEND;TZID=Asia/Seoul:20260626T113000
DTSTAMP:20260625T033224Z
CREATED:20260528T012227Z
LAST-MODIFIED:20260625T033224Z
UID:12546-1782466200-1782473400@www.ibs.re.kr
SUMMARY:Insulin resistance prediction from wearables and routine blood biomarkers - Hyunji Jeong
DESCRIPTION:In this talk\, we discuss the paper “Insulin resistance prediction from wearables and routine blood biomarkers” by Ahmed A. Metwally et al.\, Nature\, 2026. \nAbstract: \nInsulin resistance (IR)\, a primary precursor to type 2 diabetes\, is characterized by impaired insulin action in tissues1. However\, diagnostic methods remain expensive and inaccessible\, which hinders early intervention2\,3. Here we present the WEAR-ME study\, a large\, remotely conducted study of IR (n = 1\,165 participants; median body mass index (BMI) = 28 kg m−2\, median age = 45 years\, median haemoglobin A1c (HbA1c) = 5.4%) that uses time-series data from wearable devices and routine blood biomarkers to train deep neural networks against a ground-truth measure of IR (homeostatic model assessment of IR; HOMA-IR). Using a HOMA-IR cut-off of 2.9\, our multimodal model achieved robust performance (area under the receiver operating characteristic curve (AUROC) = 0.80\, sensitivity = 76%\, specificity = 84%) with data from wearable devices\, together with demographic and routine blood biomarker data. To enhance the use of time-series data from wearables\, we fine-tuned a wearable foundation model (WFM) pretrained on 40 million hours of sensor data. In an independent validation cohort (n = 72)\, a model integrating WFM-derived representations with demographic data surpassed a demographics-only baseline (AUROC = 0.75 versus 0.66). Moreover\, adding WFM-derived representations to a model with demographics\, fasting glucose and a lipid panel substantially improved performance\, compared with an identical model without data from wearables (AUROC = 0.88 versus 0.76). We integrate IR prediction into a large language model to contextualize the results and facilitate personalized recommendations. This work establishes a scalable\, accessible framework for the early detection of metabolic risk\, which could enable timely lifestyle interventions to prevent progression to type 2 diabetes.
URL:https://www.ibs.re.kr/bimag/event/learning-longitudinal-health-representations-from-ehr-and-wearable-data-hyunji-jeong/
LOCATION:B232 Seminar Room\, IBS\, 55 Expo-ro Yuseong-gu\, Daejeon\, Daejeon\, 34126\, Korea\, Republic of
CATEGORIES:Journal Club
ORGANIZER;CN="Jae Kyoung Kim":MAILTO:jaekkim@kaist.ac.kr
END:VEVENT
END:VCALENDAR