Department Seminars & Colloquia
When you're logged in, you can subscribe seminars via e-mail
In a financially globalized world emerging market countries have different attributes from developed ones. Emerging countries may not be able to enjoy such benefits as consumption smoothing, efficient investment and risk diversification as developed countries do. One example is the sovereign wealth fund, a precautionary savings fund run by government. Instead, they may suffer so-called capital inflow problem caused by massive capital inflows followed by a sudden stop and reversal. The global financial crisis which hit Korea in 2008 is a classic example. The root source of the problem is that these countries, located in the periphery of the global financial world, cannot produce safe assets. The direct implication is that emerging market countries must import safe assets often financed by current account surplus. Furthermore, the central banks in emerging market countries should ultimately bear the risk of currency and maturity mismatches of domestic residents. That is, they should serve not only a lender of last resort but as an insurer of last resort, which makes central banking in emerging market countries complicated. Considering that capital flows are procylical and emerging countries are unable to produce safe assets capital flows to/from emerging market countries even strengthen the case for procylicality. Consequently, it is inevitable that emerging economies in the periphery are heavily influenced by the monetary policy of the center countries, and have essentially no room for independent monetary policy. In fact, the trilemma (inability to choose flexible exchange rates, capital mobility and independent monetary policy) in theory turns out a dilemma in reality. This seminar has two parts, central banking in emerging market and central banking in Korea. After reviewing central banking in emerging market in general I will discuss specific issues related to the monetary policy of the Bank of Korea.
The Langlands program consists of a huge web of tantalizing conjectures in many different directions, e.g., relating number theory, representation theory, algebraic geometry, and harmonic analysis in an unexpected way. After a leisurely introduction to the program through a brief historical review, we introduce yet another point of view on the program, inspired by work of physicists Kapustin and Witten, by investigating a certain 4-dimensional quantum field theory. Incidentally, this turns out to provide surprising new structures on the program. This talk, based on my thesis work, is mostly aimed at non-specialists.
The KdV equation is a nonlinear partial differential equation describing waves on shallow water surfaces. In spite of its nonlinearity, this is exactly solvable as it admits a surprisingly rich structure like infinite-dimensional symmetries, called the KdV hierarchy. From a completely different direction, for a Calabi-Yau manifold X, one can consider a generating function of certain enumerative invariants of X. Witten conjectured and Kontsevich proved the mysterious claim that when X is a point, the generating function is governed by the KdV hierarchy. I will explain a program toward understanding what happens when X is a general Calabi-Yau manifold. This talk is based on a joint project in progress with Weiqiang He and Si Li.
|
In a financially globalized world emerging market countries have different attributes from developed ones. Emerging countries may not be able to enjoy such benefits as consumption smoothing, efficient investment and risk diversification as developed countries do. One example is the sovereign wealth fund, a precautionary savings fund run by government. Instead, they may suffer so-called capital inflow problem caused by massive capital inflows followed by a sudden stop and reversal. The global financial crisis which hit Korea in 2008 is a classic example. The root source of the problem is that these countries, located in the periphery of the global financial world, cannot produce safe assets. The direct implication is that emerging market countries must import safe assets often financed by current account surplus. Furthermore, the central banks in emerging market countries should ultimately bear the risk of currency and maturity mismatches of domestic residents. That is, they should serve not only a lender of last resort but as an insurer of last resort, which makes central banking in emerging market countries complicated. Considering that capital flows are procylical and emerging countries are unable to produce safe assets capital flows to/from emerging market countries even strengthen the case for procylicality. Consequently, it is inevitable that emerging economies in the periphery are heavily influenced by the monetary policy of the center countries, and have essentially no room for independent monetary policy. In fact, the trilemma (inability to choose flexible exchange rates, capital mobility and independent monetary policy) in theory turns out a dilemma in reality. This seminar has two parts, central banking in emerging market and central banking in Korea. After reviewing central banking in emerging market in general I will discuss specific issues related to the monetary policy of the Bank of Korea.
In this talk, I will describe a project (work in progress) with Sang-Bum Yoo, on birational geometry of moduli spaces of parabolic bundles on the projective line in the framework of Mori’s program. It exhibits an interesting connection between representation theory and birational algebraic geometry via invariant theory.
논문 및 특허분석을 포함한 연결망 분석에서 시간이 가장 많이 소요되는 단계가 데이터 전처리와 행렬생성, 그리고 Pajek과 Gephi 등과 같은 네트워크 분석 전문 S/W에 적합한 입력데이터를 만드는 과정입니다. KnowledgeMatrix Plus는 1) 한글 및 영문 명사 추출, stemming, 시소러스 등을 통해 데이터 전처리를 지원하고, 2) 다양한 형태의 1-mode matrix와 2-mode matrix를 생성하여, 3) 네트워크 분석 및 가시화 S/W인 VOSviewer, Pajek, Gephi에서 즉시 분석을 할 수 있도록 지원하는 소프트웨어입니다. 이번 세미나를 통해 데이터 전처리에서 행렬생성, 네트워크 분석 및 가시화까지의 전과정을 다양한 공개 S/W를 활용하여 분석하는 방법을 함께 공유하고자 합니다.
○ 당일 세미나 자료 이외의 조금 긴 자료와 샘플데이터는,
http://mirian.kisti.re.kr/km/km.jsp 의 \'Notice - 과학기술 계량분석 공개 튜토리얼 자료집\'(압축파일)을 참조하시기 바랍니다.
○ Gephi의 경우 기존 자바와 충돌이 일어나 설치가 되어도 실행이 안 될 경우가 있습니다.
이 경우에 첨부한 gephi_jre7_지정하는 방법.pdf를 참조하시기 바랍니다.
Let M be a closed manifold that admits a nontrivial cover diffeomorphic to itself. Which manifolds have such a self-similar structure? Examples are provided by tori, in which case the covering is homotopic to a linear endomorphism. Under the assumption that all iterates of the covering of M are regular, we show that any self-cover is induced by a linear endomorphism of a torus on a quotient of the fundamental group. Under further hypotheses we show that a finite cover of M is a principal torus bundle. We use this to give an application to holomorphic self-covers of Kaehler manifolds.
Credit derivatives such as CDS and CDOs are financial instruments to hedge against the default risk. CDS is a kind of insurance that compensates for the loss of bonds issued by an individual company. CDOs issue a new security called tranche, which has various types of profit and risk structures. This lecture covers the structure and pricing of CDS and CDOs.
In order to measure the return or risk of an investment, it is necessary to reflect the correlation between the return of investment assets and the default random variables. Traditionally, correlation coefficients explain the dependency structure between elliptic distributions such as normal distribution well, but new dependency measures are needed because the actual data show a much different pattern from the elliptical distributions. Copulas are widely used as an alternative to these correlation coefficients and are often used in risk management and the valuation of credit derivatives such as CDOs. This lecture will cover the basic concepts of copula and its applications.
Fusion Hall(1F), KI Bldg.(#E4), KAIST, Daejeon
Distinguished Scholar Lecture Series
Terence Tao (UCLA)
Finite time blowup constructions for supercritical equations
Many basic PDE of physical interest, such as the three-dimensional Navier-Stokes equations, are "supercritical" in that the known conserved or bounded quantities for these equations allow the nonlinear components of the PDE to dominate the linear ones at fine scales. Because of this, almost none of the known methods for establishing global regularity for such equations can work, and global regularity for Navier-Stokes in particular is a notorious open problem. We present here some ways to show that if one allows some modifications to these supercritical PDE, one can in fact construct solutions that blow up in finite time (while still obeying conservation laws such as conservation of energy). This does not directly impact the global regularity question for the unmodified equations, but it does rule out some potential approaches to establish such regularity.
2017 CMC Distinguished Lecture Series by Terence Tao
E6-1, ROOM 1409
Discrete Math
Henry Liu (Central South University, Changsha, China)
Highly connected subgraphs in sparse graphs
Let G be a graph on n vertices with independence number α. How large must a k-connected subgraph G contain? We shall present the best possible answers when α=2 and α=3. Some open questions will also be presented.
Fusion Hall(1F), KI Bldg.(#E4)
Distinguished Scholar Lecture Series
Terence Tao (UCLA)
The Erdős discrepancy problem
The discrepancy of a sequence f(1), f(2), ... of numbers is defined to be the largest value of |f(d) + f(2d) + ... + f(nd)| as n,d range over the natural numbers. In the 1930s, Erdos posed the question of whether any sequence consisting only of +1 and -1 could have bounded discrepancy. In 2010, the collaborative Polymath5 project showed (among other things) that the problem could be effectively reduced to a problem involving completely multiplicative sequences. Finally, using recent breakthroughs in the asymptotics of completely multiplicative sequences by Matomaki and Radziwill, as well as a surprising application of the Shannon entropy inequalities, the Erdos discrepancy problem was solved in 2015. In this talk I will discuss this solution and its connection to the Chowla and Elliott conjectures in number theory.
2017 CMC Distinguished Lecture Series by Terence Tao
Hyperrings and hyperfields are algebraic structures which generalize commutative rings and fields. In this talk, we aim to introduce these exotic structures and also provide examples which illustrate how hyperrings and hyperfields show up in algebraic geometry and combinatorics following the idea of Baker and Bowler on `matroids over hyperfields'.
자연과학동 E6-1, ROOM 1409
Discrete Math
O-joung Kwon (Technische Universitat Berlin, Berin, Germany)
On low rank-width colorings
We introduce the concept of low rank-width colorings, generalizing the notion of low tree-depth colorings introduced by Nešetřil and Ossona de Mendez in [Grad and classes with bounded expansion I. Decompositions. EJC 2008]. We say that a class C of graphs admits low rank-width colorings if there exist functions N:ℕ→ℕ and Q:ℕ→ℕ such that for all p∈ℕ, every graph G∈C can be vertex colored with at most N(p) colors such that the union of any i≤p color classes induces a subgraph of rank-width at most Q(i).
Graph classes admitting low rank-width colorings strictly generalize graph classes admitting low tree-depth colorings and graph classes of bounded rank-width. We prove that for every graph class C of bounded expansion and every positive integer r, the class {Gr: G∈C} of r-th powers of graphs from C, as well as the classes of unit interval graphs and bipartite permutation graphs admit low rank-width colorings. All of these classes have unbounded rank-width and do not admit low tree-depth colorings. We also show that the classes of interval graphs and permutation graphs do not admit low rank-width colorings. In this talk, we provide the color refinement technique necessary to show the first result. This is joint work with Sebastian Sierbertz and Michał Pilipczuk.
In this talk, we consider stochastic partial differential equations, especially, a parabolic Anderson model. This model shows intermittent phenomena, i.e., the solution becomes very big on small regions of different scales (we say tall peaks occur on small islands). We provide a way to quantify tall peaks and small islands by using the macroscopic fractal dimension theory by Barlow and Taylor. This is based on joint work with Davar Khoshnevisan and Yimin Xiao.
The interactions between particles and uid have received a bulk of attention due to a number of their applications in the eld of, for example, biotechnology, medicine, and in the study of sedimentation phenomenon, compressibility of droplets of the spray, cooling tower plumes, and diesel engines, etc. In this talk, we present coupled kinetic- uid equations. The proposed equations consist of Vlasov-Fokker-Planck equation with local alignment forces and the incompressible Navier-Stokes equations. For the equations, we establish the global existence of weak solutions, hydrodynamic limit, and large-time behavior of solutions. We also remark on blow-up of classical solutions in the whole space.
High-frequency financial data allow us to estimate large volatility matrices with relatively short time horizon. Many novel statistical methods have been introduced to address large volatility matrix estimation problems from a high-dimensional Ito process with micro-structure noise contamination. Their asymptotic theories require sub-Gaussian or some finite high-order moments assumptions. These assumptions are at odd with the heavy tail phenomenon that is pandemic in financial stock returns and new procedures are needed to mitigate the influence of heavy tails. In this paper, we introduce the Huber loss function with a diverging threshold to develop a robust realized volatility estimation. We show that it has the sub-Gaussian concentration around the conditional expected volatility with only finite fourth moments. With the proposed robust estimator as input, we further regularize it by using the principal orthogonal component thresholding (POET) procedure to estimate the large volatility matrix that admits an approximate factor structure. We establish the asymptotic theories for such low-rank plus sparse matrices. The simulation study is conducted to check the finite sample performance of the proposed estimation methods.
2017 제4회 정오의 수학산책
강연자: 양성덕 (고려대)
일시: 2017년 6월 2일(금) 12:00 ~ 13:15
장소: 카이스트 자연과학동 E6-1 3435호
제목: 고차원 공간의 계량과 곡률
내용: TBA
등록: 2017년 5월 31일(수) 오후 3시까지
https://goo.gl/forms/ol819KdGNA5SdGSr2
The circle is the only connected closed 1-dimensional manifold, and maybe that's why it has so many interesting features. In this talk, we would like to emphasize that there are many things we still do not know about this one of the simplest manifolds. We will survey many interesting recent results around the circle in the context of low-dimensional topology.
We consider the incompressible Navier-Stokes equations ((NS) in short) in $mathbb{R}^{3}$.
We consider the incompressible Navier-Stokes equations ((NS) in short) in $mathbb{R}^{3}$.
We consider the incompressible Navier-Stokes equations ((NS) in short) in $mathbb{R}^{3}$.
In the talk, I discuss previous works on the arithmetic of various twisted special $L$-values and dynamical phenomena behind them. Main emphasis will be put on the problem of estimating several exponential sums such as Kloosterman sums and its relation to the problem of non-vanishing of special $L$-values with cyclotomic twists. A distribution of homological cycles on the modular curves will also be discussed and as a consequence, some results on a conjecture of Mazur-Rubin-Stein about the distribution of period integrals of elliptic modular forms will be presented.
We show that the bipolar filtration of the smooth concordance group of topologically slice knots introduced by Cochran, Harvey and Horn has nontrivial graded quotients at every stage. To detect a nontrivial element in the quotient, the proof uses Cheeger-Gromov $L^2$ $rho$-invariants and infinitely many Heegaard Floer correction term invariants simultaneously. This is joint work with Jae Choon Cha.
The standard problem of optical tomography is to obtain information about the optical parameters inside of an object by making optical measurements on the boundary. Acousto-optic tomography is a variation of this problem where the object is perturbed by an acoustic field, and optical boundary measurements are taken as the parameters of the acoustic field vary. In this talk I will give a short introduction to the idea of acousto-optic tomography, and discuss some inverse problems that arise from this imaging technique. In particular, I will describe some recent results for inverse problems derived from radiative transport models of acousto-optic tomography. This is joint work with John Schotland and Guillaume Bal.
We discuss the asymptotic behavior, at a small viscosity, of solutions to some fluid equations related to the Navier-Stokes equations. The model equations, we consider in this talk, are either supplemented with the Navier-slip type boundary conditions, or simplified under some special symmetries or linearized when the no-slip boundary condition is imposed. By explicitly constructing the boundary layer correctors, which approximate the differences between the viscous and inviscid solutions, we validate the smallness of our asymptotic expansions with respect to the viscosity parameter, and prove the vanishing viscosity limit with the optimal rates of convergence.
The aim is to introduce recent progress on partial regularity problem of the Navier-Stokes equations.
They include an improved regularity criteria and the size of possible singularities in terms of the Minkowski dimension.
We also discuss about the regularity and singularity properties of the weak solutions to the Navier-Stokes equations which belong to some weak Lebesgue spaces.
The motivation for the talk is the recent result, joint with Vincent Guirardel and Camille Horbez, that Out(F_n) admits a topologically amenable action on a Cantor set. This implies the Novikov conjecture for Out(F_n) and its subgroups. Most of the talk will be an introduction to boundary amenability and ways to prove it for simpler groups.
2017 제3회 정오의 수학산책
강연자: 이수준(경희대)
일시: 2017년 5월 12일(금) 12:00 ~ 13:15
장소: 카이스트 자연과학동 E6-1 3435호
제목: 양자정보이론 소개 : 디지털정보 vs 양자정보
내용: 이 강연에서는, 디지털정보의 데이터 압축, 오류 보정에 대한 C. E. Shannon의 두 가지 중요한 정보이론의 결과를 설명하고, 이와 함께 양자정보에 대한 정보이론과의 근본적인 차이점에 대해서 소개하고자 한다.
등록: 2017년 5월 10일(수) 오후 3시까지
https://goo.gl/forms/TnDIENKzVJAnxy9v2
The motivation for the talk is the recent result, joint with Vincent Guirardel and Camille Horbez, that Out(F_n) admits a topologically amenable action on a Cantor set. This implies the Novikov conjecture for Out(F_n) and its subgroups. Most of the talk will be an introduction to boundary amenability and ways to prove it for simpler groups.
This talk will review previous work on quadrupedal gaits and recent work on a generalized model for binocular rivalry proposed by Hugh Wilson. Both applications show how rigid phase-shift synchrony in periodic solutions of coupled systems of differential equations can help understand high level collective behavior in the nervous system.
Probabilistic methods in geometric group theory have recently gained in importance. In the course I will focus on the setting where a countable group G acts on a (possibly nonproper) Gromov hyperbolic space X. Examples include a mapping class group acting on the associated curve complex, or Out(F_n) acting on the complex of free factors, or a group acting on the contact graph of a CAT(0) cube complex it acts on. Under minor assumptions, Maher-Tiozzo show that a random walk on G, projected to X, almost surely converges to a point in the Gromov boundary of X. I will discuss the proof of this theorem. As an application, we will see that "generic" elements of mapping class groups are pseudo-Anosov, and (following Horbez) we will give a random walk proof of the classical theorem of Ivanov classifying subgroups of mapping class groups.
Probabilistic methods in geometric group theory have recently gained in importance. In the course I will focus on the setting where a countable group G acts on a (possibly nonproper) Gromov hyperbolic space X. Examples include a mapping class group acting on the associated curve complex, or Out(F_n) acting on the complex of free factors, or a group acting on the contact graph of a CAT(0) cube complex it acts on. Under minor assumptions, Maher-Tiozzo show that a random walk on G, projected to X, almost surely converges to a point in the Gromov boundary of X. I will discuss the proof of this theorem. As an application, we will see that "generic" elements of mapping class groups are pseudo-Anosov, and (following Horbez) we will give a random walk proof of the classical theorem of Ivanov classifying subgroups of mapping class groups.
Probabilistic methods in geometric group theory have recently gained in importance. In the course I will focus on the setting where a countable group G acts on a (possibly nonproper) Gromov hyperbolic space X. Examples include a mapping class group acting on the associated curve complex, or Out(F_n) acting on the complex of free factors, or a group acting on the contact graph of a CAT(0) cube complex it acts on. Under minor assumptions, Maher-Tiozzo show that a random walk on G, projected to X, almost surely converges to a point in the Gromov boundary of X. I will discuss the proof of this theorem. As an application, we will see that "generic" elements of mapping class groups are pseudo-Anosov, and (following Horbez) we will give a random walk proof of the classical theorem of Ivanov classifying subgroups of mapping class groups.
We consider non-topological solutions of a nonlinear elliptic system problem derived from the SU(3) Chern-Simons models.
The existence of non-topological solutions even for radial symmetric case has been a long standing open problem.
Recently, Choe, Kim, and Lin showed the existence of radial symmetric non-topological solution when the vortex points collapse.
However, the arguments in that paper cannot work for an arbitrary configuration of vortex points.
In this talk, I introduce a new approach by using different scalings for different components of the system to construct a family of partial blowing up non-topological solutions.
This talk is based on the joint work with Prof. Chang-Shou Lin and Prof. Ting-Jung Kuo.
In this talk we discuss the generation of interface property for solutions of the nonlocal Allen-Cahn equation which was proposed by Rubinstein and Sternberg as a model for phase separation in a binary mixture. More precisely,
we show that given an arbitrarily initial condition, the solution approaches a step function and hence develops a steep transition layer (interface) within a very short time. Because of the nonlocal term, some PDE tools such as comparison principle cannot be applied so that we have to introduce new method to overcome these diculties. Furthermore, in some cases, we obtain a sharp estimate for the thickness of interface.
This is joint work with Danielle Hilhorst, Hiroshi Matano and Hendrik Weber.