MCIM students will give talks on their internships and research on the Fridays when there are no outside industrial speakers scheduled. If you would like to sign up to give a talk, or make an appointment to speak with any of the industrial speakers listed below please call 612-625-3377 or e-mail Rhonda

Fall 2002

• September 27, 10:10am, 570 VinH
  Anton Leykin, School of Mathematics, University of Minnesota
  2-dim guillotine cutting problem: 3 different approaches
  

• October 4, 2002, 10:10am, 570 VinH
  Dr. Ron Mahler, Lockheed Martin Tactical Defense Systems
  Tracking in High Target Densities Using a First-Order
  Multitarget Moment Density
  

• October 11, 2002, 10:10am, 570 VinH
  Dr. Fred Hulting, General Mills
  Statistics in New Product Development

• November 1, 2002, 10:10am, 570 VinH
  Oleg Aleksandrov (Math Grad Student)
  

• November 22, 2002, 10:10am, 570 VinH
  Dr. Mihalis Sigalas, Agilent Technology

  Photonic Crystals in Optical Communications
  Recent trends in optical communications show an increase in device integration along with a decrease in device size.  Photonic crystals (PC) may be the platform of future miniaturize optical devices because they can control the light in sizes of the order of the wavelength.   The theoretical tools needed to study PC will be presented.   Results for both two dimensional slab PC and three dimensional PC will be shown and the advantages of each case will be discussed.

• December 13, 2002, 10:10 am, 570 VinH
  Dr. Ann Dewitt, 3M

  Mathematics Applied to Biological Systems in Drug Discovery
  Advances in tools to probe biological phenomena such as combinatorial chemistry, high-throughput screening, genomics and proteomics have, in part, resulted in a rapid rise in the rate at which information is collected. The corresponding increase in the volume of information supplies a rich source for understanding how biological systems operate, but appropriate methods for placing each new piece of information into a larger context must be developed. Certainly mathematics have been applied to the investigation of biological systems in the past, and further opportunities arise from the need to organize and understand vast amounts of information, and to, furthermore, systematically, quantitatively capture behavior for predictive engineering. This presentation will focus on how mathematics is used as a data analysis and predictive engineering tool to understand biological processes (i.e. life!), including a general introduction to the emerging discipline of "systems biology." Doctoral research conducted at Massachusetts Institute of Technology will be used for illustration along with examples from current research conducted in 3M Pharmaceuticals.
  

Spring 2003

• January 31, 2003,10:10am, 570 VinH
  Stephen Mildenhall, Kemper Insurance

• February 7, 2003, 10:10am, 570 VinH
  Lawrence Cowsar, Lucent Technologies

• February 21, 2003, 10:10am, 570 VinH
  Lili Ju, IMA Industrial Postdoc
  
  Cortical Surface Flattening Using Discrete Conformal Mapping with Minimal Metric Distortion
  Although flattening a cortical surface necessarily introduces metric distortion due to the non-constant Gaussian curvature of the surface, the Riemann Mapping Theorem states that continuously differentiable surfaces can be mapped without angular distortion. Several techniques have been proposed for flattening polygonal representations of surfaces while substantially minimizing metric distortion, and methods for conformal flattening of polygonal surfaces have also been proposed. We describe
  an efficient method for generating conformal flat maps of triangulated surfaces while minimizing metric distortion within the class of conformal maps. Our method, which controls both angular and metric distortion, involves the solution of a linear system and a small scale nonlinear minimization. It can be applied to user-defined "patches" or to an entire
  cortical surface.

• February 28, 2003, 10:10am, 570 VinH
  TBA

• March 7, 2003, 10:10am, 570 VinH
  Kevin Ellwood, Ford

• April 18, 2003, 10:10am, 570 VinH
  Richard Chiao, GE Medical
  
  Diagnostic Ultrasound: Technology and Applications
  Ultrasound has developed over the past 50 years into a major diagnostic imaging modality, complementing CT, MRI and nuclear imaging. Major applications of ultrasound today include cardiovascular, abdominal organs, muskloskeletal, small parts, and OB/Gyn. Increased clinical usage of ultrasound has been driven by technological advances that exploit the following advantages compared to other modalities: real-time (especially important for heart and blood flow), safe due to non-ionizing radiation, portable, and low cost. Basic ultrasound modes include B-mode that images the acoustic reflectivity of tissue structures and Doppler that measures blood velocity. Recent advances include harmonic imaging that improves image quality by exploiting the nonlinear behavior of high-amplitude ultrasound propagation in tissue or micro-bubble contrast agents, and code technology that circumvents traditional resolution / penetration tradeoffs. Future directions for ultrasound are at the intersection of clinical needs (image quality, new applications, and increased productivity) and major technological trends (miniaturization, SW), which include miniaturized systems and probe components, improved image quality, new imaging parameters, and 4D imaging.

• April 25, 2003, 10:10am, 570 VinH
  Nicholas Bennett, Schlumberger Doll Research
  
  Posterior Uncertainty in Decimated Wavelet Model Parameterizations
  Solving a geophysical inverse problem means determining the parameters of an earth model given a set of measurements. In solving many practicalinverse problems, accounting for the uncertainty of the solution is very important to aid in decision-making. In this work, we address the problem of determining the posterior uncertainty of the solution for models that arise from decimated wavelet bases using a simple 1 dimensional seismic travel time inversion problem. Our inversion methodology is to pick a model decimation, prepare a prior mean and covariance matrix of the wavelet coefficients, compute a posterior mean and covariance, and then to sample from this posterior distribution. We also sample different choices of model decimation in proportion to their posterior probability. These samples span the uncertainty of the inverse problem solution, accounting for both the uncertainty in the choice of model decimation and of wavelet coefficients. We note that a re-normalization of the decimated prior covariance matrix of the wavelet coefficients is required to properly account for the amount of variance in the prior distribution. Further, we present a fast algorithm for computing this normalized decimated prior covariance matrix.

• May 2, 2003, 10:10am, 570 VinH
  Daniel Baker, General Motors R&D Center

  Impedance as a diagnostic tool for studying fuel cells
  We will start by showing some CFD simulations of current distribution on a fuel cell. The current distribution behaves differently under different operating conditions and we will try to explain this behavior. This will lead us to consider impedance spectroscopy as a tool to investigate some of the critical effects that impact current distribution. A short explanation of impedance methods will be given along with a discussion of how to interpret impedance data in the context of current distribution. Special emphasis will be given to the high frequency resistance (HFR) as a tool for understanding membrane humidification. Other impedance applications include assessing proton resistance in the porous cathode, kinetic resistance of the cathode electrode, and the relative contribution of gas transport resistance to voltage losses.

• May 13, 3002 2:30 pm, 409 Lind Hall
  Andrew Mullhaupt, S.A.C. Capital Management

  Cantelli’s Lemma and the Estimation of Transaction Costs
  There is a great variety of mathematics that has found its way into the world of finance. Without a doubt a great deal of this work occurs at hedge funds and investment companies who profit directly from the use of mathematics. One area of particular interest is the mathematical study of transaction costs – those costs associate with buying and selling in financial markets. The estimation of such costs have interesting – and sometimes surprising – mathematical limits which allow us to illustrate in a general sense the flavor of the mathematical research that takes place within our group. When a trader buys or sells in financial markets the trade aects the very market in which the transaction takes place. Mathematical methods for depend
  on our ability to bound what is – and especially what is not – within the realm of possibility.
  In this talk we give an introduction to the mathematics and economics of transaction costs then show how Cantelli’s lemma may be used to make some surprising statements about the limits of our ability to estimate transaction costs. We also give a simple derivation of Cantelli’s lemma, which states that for a random variable X with mean µ and standard deviation s  we have

Finally, we give applications of the above to portfolio selection theory.

About the Speaker

Andrew Mullhaupt is Director of Research of the Meridian Group at S.A.C. Capital Management in New York. S.A.C. Capital Management is a Stamford and New York City based private investment firm. Andrew has a PhD in Mathematics from the Courant Institute of Mathematical Sciences and has held positions at Morgan Stanley and Renaissance Technologies.

2001-2002 Industrial Problems Seminar