1. 일시: 2023년 4월 12일(수), 14:00-16:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

   • 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
   • 무료주차는 2시간 지원됩니다.

3. 발표자: 최호식 교수(서울시립대학교)

4. 주요내용: Quantile Estimation for Encrypted Data

Third-party analyzer which uses personal information may expose sensitive information. As data-based researches increase, such privacy issues are being raised seriously. Homomorphic encryption(HE) has been proposed as a way to avoid it. However, a challenge in ciphertext used in HE is a calculation time of elementary operations, which has significantly much higher complexity than that on plaintexts, resulting in that subsequent data analysis limiting various statistical analytics compared to plaintext data. In this paper, we consider an estimation method to get quantiles for encrypted data, where quantiles are core statistics to figure out data distribution in statistical data analysis. We propose a HE-friendly algorithm for large homomorphic encrypted data using an approximate quantile loss function. Numerical studies show that our proposed method improves calculation time over an estimation based on a sorting way for randomly generated and real data's homomorphically encrypted data. Furthermore, we suggest a modified boxplot for homomorphically encrypted data, where a boxplot is a primary statistical analysis method necessary for data analysis.

유튜브 실시간 스트리밍 : 현장 참석이 어려운 분들을 위해 온라인으로 실시간 방송할 예정입니다. 주소는 당일 신청 페이지에 업데이트 하겠습니다.

1. 일시: 2023년 3월 16일(목), 14:00-16:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

   • 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
   • 무료주차는 2시간 지원됩니다.

3. 발표자: 장봉수 교수(울산과학기술원)

4. 주요내용: Biodiversity with RPS game and A.I

Evolutionary games of cyclic competitions have been extensively studied to gain insights into one of the most fundamental phenomena in nature: biodiversity that seems to be excluded by the principle of natural selection. The Rock-Paper-Scissors (RPS) game of three species and its extensions [e.g., the Rock-Paper-Scissors-Lizard-Spock (RPSLS) game] are paradigmatic models in this field. In this presentation, we introduce the basic mathematical models and their effects on coexistence with intrinsic symmetry and nonuniform intraspecific competitions through stability analysis, Monte Carlo simulations, and continuous spatiotemporal dynamical evolution from partial differential equations. Moreover, we highlight several exciting investigations of biodiversity considering the heterogeneity of species' features and the structural clustering shapes. Focusing on the remarkably expansive computational costs of analyzing evolutionary dynamics as increasing species populations, we introduce AI models to investigate biodiversity in RPS games.

** 유튜브 실시간 스트리밍 : 현장 참석이 어려운 분들을 위해 온라인으로 실시간 방송할 예정입니다. 주소는 당일 신청페이지에 업데이트 하겠습니다.

1. 일시: 2023년 3월 7일(화), 10:00-14:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

   • 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
   • 무료주차는 2시간 지원됩니다.

3. 발표자 및 주요 내용

10:00-11:30

• 연사: 지종호 박사(아주대학교 의료원)
• 주제: Grap Representation Learning in Biological and Social Networks
• 초록: Graph data is used in a wide range of domains, such as biominformatics, pharmacology, and social networks, and network security. In bioinformatics, a set of genes—referred to as a genome—is usually represented as a graph, with nodes denoting genes and edges the relations between them. Or people are nodes and interaction between them are edges in social networks. Representative tasks using these graphs are finding key genes and side effects of drugs in the biological network, or finding crime suspects in the crime network (node/edge classification, etc.). Graph representation learning, has become more important in scientific and medical domains. Graph learning in these domains is to discover distinguishing features of chemical compounds or groups of people. This talk is to introduce some of machine learning algorithms such as graph neural networks and graph-based semi-supervised learning in application to bio/social networks.

12:30-14:00

• 연사: 이수연 박사(아주대학교 의과대학)
• 주제: Introduction of diverse bio-medical approaches to conducting drug repositioning and therapeutic target discovery based on multi-omics data and deep learning and statistical approaches.
• 초록: Various deep learning and statistical approaches have been applied to big multi-omics data of cancer patients are being applied to identify biomarkers and new drug efficacy discovery of diverse cancer types these days. However, multi-omics data generally have a character with high dimensions compared with relatively few patient samples, this imbalance is a recognized bottleneck to applying integrated characteristics of multi-omics in bio-clinical research. For these limitations, Diverse multi-omics data integration approaches have been applied to predict optimized potential anti-cancer and disease therapeutic target genes and drugs in diverse research. This talk will introduce kinds of applications and methods to discover new efficacy based on drug repositioning theory and cancer multi-omics data, and diverse projects that have been performed in KIURI bio-AI center on the basis of diverse bio and medical data.

현장 강의만 진행합니다. 식사 제공 없습니다

1. 일시: 2023년 3월 6일(월), 14:00-16:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

   • 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
   • 무료주차는 2시간 지원됩니다.

3. 발표자: 노영균 교수(한양대학교)

4. 주요내용: Techniques for f-divergence Estimation and their Applications to Machine Learning

   In this talk, various f-divergence estimation techniques and their applications will be explained based on the connection to loss functions. It is known that loss functions are connected to their corresponding f-divergences which measure the theoretical limit of the associated loss functions. In this regard, researchers have focused on the problems that tackle the optimization of f-divergences instead of loss functions. I will clarify the problems that we can benefit from optimizing the f-divergences and explain how novel techniques for f-divergence estimation can be used for real applications.

1. 일시: 2023년 2월 27일(월), 14:00-16:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

   • 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
   • 무료주차는 2시간 지원됩니다.

3. 발표자: 권순학 교수(성균관대학교)

4. 주요내용: Survey of vectorial Boolean functions and some unsolved problems.

Boolean functions have many applications in diverse areas of mathematics and engineering, and an efficient design of S-Box in block ciphers is one such example. It is desirable to construct invertible Boolean functions (or a class of permutation polynomials over finite fields) having good cryptographic parameters. In this talk, we give a brief survey of (vectorial) Boolean functions and their applications. We will explain standard invariants of Boolean functions such as nonlinearity, differential uniformity, Walsh spectrum, etc. Also, some results and conjectures on the functions having optimal invariants will be discussed.

** 유튜브 실시간 스트리밍 : 현장 참석이 어려운 분들을 위해 온라인으로 실시간 방송할 예정입니다. 주소는 당일 신청페이지에 업데이트 하겠습니다.

1. 일시: 2023년 2월 23일(목), 11:00-13:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

   • 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
   • 무료주차는 2시간 지원됩니다.

3. 발표자: 권순선 교수(아주대학교)

4. 주요내용: Interpretable Classification for Multivariate Gait Analysis

Motivated by gait data from both the normal and the cerebral palsy (CP) patients group with various gross motor function classification system (GMFCS) levels, we propose a multivariate functional classification method to investigate the relationship between kinematic gait measures and GMFCS levels. A sparse linear functional discrimination framework is utilized to achieve an interpretable prediction model. The method is generalized to handle multivariate functional data and multi-class classification. The method yields superior prediction accuracy and provides easily interpretable discriminant functions. And it will help clinicians to diagnose CP and assign an appropriate GMFCS level in a more consistent and mathematical evidence.

** 유튜브 실시간 스트리밍 : 현장 참석이 어려운 분들을 위해 온라인으로 실시간 방송할 예정입니다. 주소는 당일 신청페이지에 업데이트 하겠습니다.

인공지능에 관심 있는 수학자 및 관련 연구자들을 대상으로 최근 주목을 끌고 있는 GNN(Graph Neural Network), Neural ODE, 그리고 Score-based 생 성모델 및 Diffusion 확률모델을 소개하고 관련 연구 정보를 제공하고자 튜토 리얼 워크숍을 개최합니다.

수학 및 관련 연구자들이 인공지능 전문가가 아니더라도 관련 주제에 손쉽게 접근할 수 있도록 강연과 함께 컴퓨팅 시연을 포함한 튜토리얼 형식의 프로그램이니 많은 참여 바랍니다.

등록 링크: https://bit.ly/2023-winter-ai-tutorial

1. 일시: 2023년 2월 16일(목), 14:00-16:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

   • 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
   • 무료주차는 2시간 지원됩니다.

3. 발표자: 정은옥 교수(건국대학교)

4. 주요내용: Lessons from COVID-19 and Mathematical Modeling

Due to COVID-19, our society, culture, economy, public health and education have changed a lot. In the mathematics community, research has also been conducted to present a preemptive and scientific basis for the COVID-19 quarantine policy through communication with the Korea Disease Control and Prevention Agency (KDCA) rather than existing laboratory research. In this talk, we will present some mathematical modeling that supported the disease control policy decision of the KDCA during COVID-19 epidemics. Furthermore, let’s think together what mathematicians should do with COVID-19 and what the important lessons from COVID-19 are.

** 유튜브 실시간 스트리밍 : 현장 참석이 어려운 분들을 위해 온라인으로 실시간 방송할 예정입니다. 주소는 당일 신청 페이지에 업데이트 하겠습니다.

1. 일시: 2023년 2월 3일(금), 15:00-17:00

2. 장소: 산업수학혁신센터 세미나실

• 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
• 무료주차 2시간 등록 가능

  3. 발표자: 김연응 교수(가천대학교)

  4. 주요내용: Discrete optimal control - learning LQR

  Linear Quadratic Regulator (LQR) is one of the most popular structures in optimal control problems. This talk introduces some basic properties of the LQR problems and recent progress in learning LQR. There are two approaches to measuring the performance of the algorithm, Bayesian and Frequentist regret. The advantages and disadvantages will be discussed particularly g on Thompson sampling. Thompson sampling (TS) is known to effectively address the exploration-exploitation trade-off in online learning problems including reinforcement learning for linear-quadratic regulators (LQR). However, in TS for learning LQR, its theoretical analysis is often limited to the case of Gaussian noises. The sampling can be performed directly when we further assume that the unknown system parameters lie in a prespecified compact set which is seemingly restrictive. We propose a new TS algorithm for LQR, exploiting Langevin dynamics to handle a larger class of problems including those with non-Gaussian noises. The notion of preconditioner is introduced to generate samples from non-conjugate posterior distributions.

현장강의만 진행합니다.

1. 일시: 2023년 1월 19일(목), 14:00-16:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

- 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소

- 무료주차는 2시간 지원됩니다.

3. 발표자: 최수영 교수(아주대학교)

4. 주요내용: GPU-friendly enumeration for mathematical algorithm on simplicial complexes

In this talk, we discuss the use of parallel computing for mathematical problems. In particular, we provide a GPU-friendly algorithm for obtaining all weak pseudo-manifolds whose facets are all in an input set of facets satisfying given conditions.

We use this algorithm to completely list up toric colorable seed PL-spheres with a few vertices implying the complete characterization of PL-spheres of dimension n-1 with n+4 vertices having maximal Buchstaber numbers. This result is an essential intermediate step for the classification of non-singular complete toric varieties with Picard number 4.

5. 유튜브 실시간 스트리밍 : 현장 참석이 어려운 분들을 위해 온라인으로 실시간 방송할 예정입니다. 주소는 당일 신청페이지에 업데이트 하겠습니다.