| 以文本方式查看主题 - SCI论坛 (http://artsoncqu.eicp.top/scibbs/index.asp) -- 最新动态 (http://artsoncqu.eicp.top/scibbs/list.asp?boardid=37) ---- 中国工业与应用数学学会第十届学术年会 (http://artsoncqu.eicp.top/scibbs/dispbbs.asp?boardid=37&id=9878) |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| -- 作者:sciadmin -- 发布时间:2008/8/20 16:18:03 -- 中国工业与应用数学学会第十届学术年会 “中国工业与应用数学学会第五届代表大会和第十届年会” 暨“第二届CSIAM苏步青应用数学奖颁奖大会”
问题驱动的应用数学 Problem-Oriented Applied Mathematics Zhengzhou University, Henan, China August 27-31, 2008 PROGRAM 中国工业与应用数学学会代表大会 (地点:嵩山饭店 具体地点?) August 26, 2008 Evening 20:00—21:30 主持人:李大潜 复旦大学 中国工业与应用数学学会第四届理事会全体会议 August 27, 2008 Morning 8:30—12:20 主持人:郭雷 中国科学院 1. 开幕式 Opening Ceremony (1) 地方官员致词 (2) 理事长李大潜院士致词 (3) 外宾致词 2. 第二届CSIAM苏步青应用数学奖(CSIAM Su Buchin Applied Mathematics Prize)颁奖 (1) 宣读获奖名单并介绍评奖程序 (2) 获奖人介绍 3. 照相 Group Photo 4. 中国工业与应用数学学会第五届代表大会(所有与会代表参加) (1) 全体代表投票选举中国工业与应用数学学会第五届理事会 (2) 第四届理事会工作报告 (3) 第四届理事会财务报告 (4) 宣布选举结果 5. 中国工业与应用数学学会第五届理事会第一次会议,投票选举常务理事 12:30 Lunch August 27, 2008 Evening 20:00—21:00 主持人:郭雷 中国科学院 中国工业与应用数学学会第五届常务理事第一次会议,投票选举理事长、副理事长和秘书长,并宣布选举结果 “中国工业与应用数学学会第五届代表大会和第十届年会” 暨“第二届CSIAM苏步青应用数学奖颁奖大会”
问题驱动的应用数学 Problem-Oriented Applied Mathematics Zhengzhou University, Henan, China August 27-31, 2008 PROGRAM 大会邀请报告 Plenary Talks (地点:嵩山饭店 具体地点?)
15:00—15:50 徐宗本 西安交通大学 基于认知的数据建模 16:00—16:50 张关泉 中国科学院 单程波方程及其在地震勘探中的应用(地下构造成像和反射系数反演) 18:00 Supper
08:30—09:20 John Ockendon OCIAM Mathematical Institute, UK Diffraction problems in Industry 09:30—10:20 Heinz W. Engl Austrian Academy of Sciences, Austria Regularization of Inverse Problems: Convergence Analysis, New Applications and Challenges 10:20—10:40 Break 10:40—11:30 曹礼群 中国科学院 Scale Gap, Scale Invariance and Mathematical Modeling at Nano-scale 11:40 Lunch 18:00 Supper 注:下午为五个分组报告(Section Talks),晚上为圆桌会议(Round Tables)
08:00—08:50 韩东 上海交通大学 统计过程控制的若干数学问题研究 08:50—09:40 徐昆 香港科技大学 Gas Dynamic Equations with Multiple Temperature 多温度非平衡态气体动力学方程的构造 09:40—10:00 Break 10:00—11:10 Huaxiong Huang MITACS, Canada MITACS, Study Groups and Problem Driven Applied Mathematics 11:10—12:00 耿直 北京大学 12:10 Lunch 18:00 Supper 注:下午为自由活动(Free)
19:30—20:20 David Cai New York University, USA Computational Modeling and Mathematical Analysis of the Dynamics of the Primary Visual Cortex 20:20—21:10 王世全 香港城市大学 Boundary Layer Theory and Exponential Asymptotics 21:10—21:20 闭幕式 PROGRAM 圆桌会议 Round Tables August 28, 2008 Evening 20:00—21:30 主题一:应用数学的教学改革 地点:嵩山饭店 具体地点? 主持人:谭永基 复旦大学 主题二:问题驱动的应用数学 地点:嵩山饭店 具体地点? 主持人:陈叔平 贵州大学 PROGRAM 分组报告 Section Talks Afternoon August 28, 2008 第一分会场:微分方程、控制理论及其应用
第二分会场:数值计算及其应用
第三分会场:数学建模
第四分会场:数理统计及综合数学应用
第五分会场:运筹与优化
基于认知的数据建模徐宗本(西安交通大学) 内容摘要:数据建模是信息技术的共有基础,是当今信息化社会数学应用的主要形式之一,其目的是揭示数据中所隐含的信息(结构、模式与规律等)。传统的数据建模方法(如统计学方法、人工智能方法)主要基于数据结构及产生数据的物理原理,我们提出基于人的认知进行数据建模的原理与方法。本报告聚焦于介绍作者在基于视感知原理进行数据建模方面所作的探索,主要包括:(1)基于视网膜多尺度空间表示的聚类、分类、回归与模型选择;(2)基于视皮层感受野机制的数据挖掘;(3)基于视觉系统神经元非线性编码机制的机器学习;(4)基于视觉感知的数据可视化等。对每一主题,我们介绍所取得的代表性成果、科学意义及其应用效果,特别地,我们展示:在每一情况下,基于视感知的数据建模常产生解决同类问题的最好算法(或最好算法之一)。所获得的结果对于推动统计学、认知科学(特别是神经科学与心理学)与信息科学的交叉有重要意义,对于解决数据挖掘、图像处理等信息技术关键问题有重要价值。 单程波方程及其在地震勘探中的应用(地下构造成像和反射系数反演)张关泉(中科院计算数学与科学工程计算研究所) 内容摘要:本报告主要汇报本人的有关研究成果及其应用情况。首先概要介绍油气地震勘探的概念,从而引出单程波(上行波,下行波)的物理含义。在一维情况下将波动算子因子分解,使整个声波波场分裂为上行波场和下行波场,并导出与声波方程等价的耦合单程波方程组,并解释其各项物理意义,然后利用平方根算子(拟微分算子)将其推广到三维的情况。为了实践中可以用于计算,将平方根算子表达为微分-积分算子,由此导出完全适用于科学计算的微分-积分方程组形式的三维耦合单程波方程组,并对其各项的物理意义作出解释和定出相应的定解条件。 报告第二部分主要介绍如何利用耦合单程波方程组处理油气勘探地震资料,使地下构造成像,并指出地下界面的反射系数如何确定的途径。最后显示几个理论数据和实际数据处理的结果。 本报告主要突出如何利用数学工具,对实际应用问题进行应用基础研究,在数学上为其奠定理论基础,并由此发展相应的数值方法,设计高效实用的算法,解决实际生产问题。 Regularization of Inverse Problems:Convergence Analysis, New Applications and ChallengesHeinz W. Engl(Austrian Academy of Sciences, Austria) Abstract: We illustrate, via parameter identification problems from systems biology, the numerical difficulties arising when solving inverse problems and give an overview over different regularization methods (variational and iterative) for their stable solution. we describe mathematical techniques for their converence analysis, where recent emphasis has been on nonlinear problems in a non-hilbert-space setting. we close by mentioning the application of such methods to inverse problems in finance and by, time permitting, a new framework for analysing stochastic inverse problems using the prokhorov metric. 统计过程控制的若干数学问题研究韩东 (上海交通大学数学系) 宗福季(香港科技大学工业工程与物流管理系) 内容摘要:统计过程控制(Statistical Process Control) 是指应用统计分析技术对生产过程进行实时监控, 科学地区分出生产过程中产品质量的随机波动与异常波动, 从而对生产过程的异常变化的时间、大小及趋势提出预警, 以便管理人员能及时采取措施, 消除异常, 恢复过程的稳定, 进而达到控制和提高产品质量的目的。 统计过程控制的理论研究虽然是从生产过程的监控开始的, 但其研究成果已广泛应用于设计、销售、服务、管理等过程。 控制图理论是研究统计过程控制的理论基础。本文将围绕控制图理论的几个基本问题: 什么样的控制图是最优的? 如何设计最优的控制图? 如何比较和判定控制图的优劣? 怎样构造更有效的控制图? 介绍我们近几年来的研究工作。
多温度非平衡态气体动力学方程的构造(Gas Dynamic Equations with Multiple Temperature)徐昆(香港科技大学) Abstract: The Bhatnagar-Gross-Krook (BGK) model approximates the collision term of the Boltzmann equation by a single relaxation process from non-equilibrium state to equilibrium one. From the BGK model, the Navier-Stokes, Burnett and Super-Burnett equations can be derived. However, the success of these derived macroscopic equations in the description of highly non-equilibrium flow is limited. In order to extend the validity of the BGK model to cope with complicated physical relaxation processes, we propose a generalized BGK (GBGK) model, where the gas distribution function settles to an equilibrium state through multiple particle collision processes. The GBGK model can be used to capture the transitional, rotational, and vibrational non-equilibrium flow phenomena. More importantly, it may provide a theoretical framework to derive new macroscopic governing equations which are valid for the near continuum flows. Based on the GBGK model, the gas-kinetic schemes have been developed and have been successfully applied to the near continuum flow computations, such as the shock structure and non-equilibrium micro-flow computations. In this talk, we are going to analyze how to derive the new macroscopic flow equations from the GBGK model. In the new governing equations, the translational temperature is defined as a tensor instead of the conventional scalar variable. At the same time, the Navier-Stokes constitutive relationship, such as the connection between stress and strain, is fully replaced by the temperature relaxation terms. These relaxation terms recover the Navier-Stokes relationship only in the limiting case as the temperature differences are very small, i.e., the so-called equilibrium limit. The newly developed macroscopic flow equations may have significant impact on the study of hypersonic rarefied flows in aerospace industry. MITACS, Study Groups and Problem Driven Applied MathematicsHuaxiong Huang(MITACS, Canada) Abstract: In this talk I will first give a brief overview of the Mathematics of Information Technology and Complex Systems (MITACS: www.mitacs.ca), an federally funded Network of Centers of Excellences in Canada. MITACS is project (problems) driven and covers five themes. I will also give my personal perspective and present some of my recent work on modeling InSb crystal growth, which started as an Oxford style study group problem and became a two-year MITACS project. I will discuss how interesting mathematical questions can be formulated by working closely with industrial partners. 因果推断的统计方法耿直(北京大学数学科学学院) 内容摘要:Democritus, (B.C.460-B.C.370?): “I would rather discover a single causal relationship than be king of Persia.” 子曰:“知之为知之,不知为不知,是知也。” 探讨事物之间的因果关系是哲学、自然科学、社会科学、医学等几乎所有科学研究的重要目的。本论文探讨从观测数据分析因果作用和挖掘因果网络的若干统计方法: 1. 井底之蛙:介绍Simpson悖论和混杂因素,因果作用的统计推断方法; 2. 替罪羔羊:介绍“中间因素悖论”,讨论替代指标,因果路径的方法; 3. 盲人摸象:介绍因果网络结构学习的分解方法; 4. 纲举目张:讨论因果网络的主动学习的方法; 寻根问底 + 顺藤摸瓜:外部干预情况下进行预测的方法,探讨因果网络的局部学习方法。 Computational Modeling and Mathematical Analysis of the Dynamics of the Primary Visual CortexDavid Cai(New York University) Abstract: We discuss our large-scale (~1 million neurons) computational modeling of the primary visual cortex (V1). In particular, we describe network mechanisms underlying stochastic, spatiotemporal dynamics associated with spontaneous on-going activity of the V1 and the line-motion illusion --- which is the illusory motion sensation from a static cue of a flashed stationary square quickly followed by a stationary bar. Related theoretical issues, such as kinetic theory of neuronal network dynamics, will also be addressed. Furthermore, we use a new analysis of coarse-grained event-chains to demonstrate the fine discriminability of orientation of V1. Boundary Layer Theory and Exponential Asymptotics王世全(Department of Mathematics, City University of Hong Kong) Abstract: In 1904, Prandtl introduced the boundary layer method (matched asymptotics) which has become one of the most important tools in applied mathematics. Despite its usefulness, there still does not seem to have a rigorous analysis to validate the accuracy of Prandtl’s treatment. In this lecture, we will point out that when exponentially small terms are taken into consideration, errors can occur even in very well-known formulas in boundary layer theory, and present some rigorous results for two non-linear two-point boundary value problems, involving Carrier-Pearson equations.
中国工业与应用数学学会第五届代表大会 和第十届年会 暨第二届CSIAM苏步青应用数学奖颁奖大会
问题驱动的应用数学 Problem-Oriented Applied Mathematics
2008年8月27-31日 河南 郑州 August 27-31, 2008 Zhengzhou, Henan
主办单位:中国工业 与应用数学学会 协办单位:郑州大学 |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||