IMS 2005 Invited Speakers

Stephen Wolfram The Future of Computation Wolfram Research Inc Abstract: Bio: Stephen Wolfram was educated at Eton, Oxford, and Caltech, receiving his PhD in theoretical physics in 1979 at the age of 20. His early work in physics and computer science was recognized by a MacArthur award in 1981. In the early 1980s he made a series of discoveries about systems known as cellular automata, leading to numerous applications in physics, mathematics, computer science, biology, and other fields. In 1986 he founded Wolfram Research, Inc. and began the creation of Mathematica — now the world's leading software system for technical computing and symbolic programming. Over the past decade Wolfram has divided his time between leadership of his company and pursuit of basic science. Wolfram's book A NEW KIND OF SCIENCE was released on May 14, 2002, and became an instant bestseller.

Luc Barthelet Research on Humanoid Robots Senior Vice President and Excecutive Producer for Electronic Arts Email: lucb@ea.com Abstract: Bio: At Electronic Arts, Luc Barthelet helped create the development environments that have contributed to the company's successful entry and leadership position in the 16-bit videogame markets. He was instrumental in the development of persistent state worlds, including Majestic, Motor City Online, Earth and Beyond, Ultima Online and, of course, The Sims Online. At Maxis, he led product development for such blockbuster titles as The Sims 2, The Sims, SimCity 4 and SimCity 3000. Although self-taught in computing and programming, he is a graduate of Ecole Superieur des Travaux Public in Paris, where he received degrees in electrical and mechanical engineering.

Dan Dubin A Better Way to Teach Mathematical and Computational Physics Department of Physics, University of California, San Diego Email: ddubin@ucsd.edu Abstract: The standard way to teach numerical and analytical methods to undergraduate science students is to break the subjects into two separate courses, one in numerics and one in analytics. These are often taught in separate departments. This talk will outline a different approach that combines numerical and analytical methods into one course of study. At UCSD the course is taught using Mathematica, in the physics department. The advantages and pitfalls of this approach will be considered, and various examples will be discussed. Bio: After graduating from the the plasma physics program at Princeton University, Dan Dubin joined the physics department at UCSD as a postdoctoral researcher. His research in theoretical and computational plasma physics has led to investigations in several related areas, including fluid dynamics, atomic physics, and structural condensed matter physics. Dubin is now a professor at UCSD. He is a Fellow of the American Physical Society (APS) and the recipient of the APS year 2000 Excellence in Plasma Physics Award.

Sarah Flannery Collaborating using Mathematica Wolfram Research Inc Email: sarahf@wolfram.com Abstract: The soon to be released Collaboration Tool will allow Mathematica users all over the world to communicate and collaborate within this powerful program. Real-time sharing of graphics, typset mathematical equations, text, code snippets, entire Notebooks—any Mathematica expression—is achieved with the click of a button. This talk will introduce the tool and suggest a broad range of usage scenarios, covering casual users, researchers, teachers, and presenters. Bio: In June 2003 Sarah gained a BA in Computer Science from the University of Cambridge, UK. In 1999 she was a winner of both the Irish and, later, European Young Scientist of the Year for work on Cryptography. The following year the popular-science bestseller In Code: A Mathematical Journey, written with her father David, was published at the request of Profile Books Ltd., London. At present she is working for Wolfram Research.

Todd Gayley The Past, Present, and Future of Mathematica-Java Integration Director of Java Technology Wolfram Research Inc Email: tgayley@wolfram.com Abstract: J/Link, a toolkit that integrates Mathematica and Java, is now five years old. By bringing the power of Java and Mathematica together in a seamless way, J/Link has made possible many major advances in the Mathematica world. We will look back at the history of J/Link and how it has matured and become more integrated with Mathematica. We will examine some of the major products that have been built on top of J/Link, some by Wolfram Research and others by third-party developers. We will then turn to some of the ways in which Java is used in the current version of Mathematica, many of which you might not be aware. All this material demonstrates the growing importance of Java as a core component of the Mathematica system, and serves as an introduction to the main focus of the presentation, which is to examine future directions for J/Link and exciting new forms of Mathematica-Java integration. These demonstrations are experimental and do not represent announced features of any future version of Mathematica, but they do involve actual running code. We will first discuss future performance and architectural improvements in J/Link, and then move on to other uses of Java such as compilation of Mathematica programs directly into Java byte code, Java cells in the front end, and a direct embedding of a Java virtual machine in the Mathematica kernel. Embedding the JVM enables an ultra high-speed Mathematica-Java interface that avoids the use of MathLink entirely. This means that for the first time programmers can write "plugins" that run at the same speed as the kernel's own internal code. Using a new Java-kernel API, programmers might one day be able to write Java code that directly creates and manipulates Mathematica expressions in the the kernel's native representation, letting users create extensions to Mathematica with essentially the same speed, power, and expressiveness that is available to the developers of the Mathematica kernel itself. Bio: After earning a PhD in evolutionary biology and being captivated by an amazing new piece of software called Mathematica, Todd came to work for Wolfram Research. Along the way, he has worked in almost every group in the company, including Technical Support, Applications, Front End, and Kernel. For the last five years he has been Director of Java Technology, creating products like J/Link and .NET/Link and helping to advance Mathematica's integration with external programs, tools, and technologies.

Jean-Pierre Hébert Artist in Residence Kavli Institute for Theoretical Physics Email: hebert@kitp.ucsb.edu Bio: Jean-Pierre Hébert is a founder member of the Algorists, a group of computer artists who had been working independently for many years, and then formed a loose association after meeting at SIGGRAPH '95. Hébert's work demonstrates that the computer genuinely creates new visual possibilities and aesthetic enquiries, while at the same time this is an exploration driven entirely by artistic considerations. Although the bulk of Hébert's pieces are created, as with many early innovators, using a plotter, he is remarkable for the range of other media that he has experimented with. Possibly the most innovative and unusual digital art works to be found are his 'sand pieces' where a steel ball is driven under algorithmic control through sand, reminiscent of Zen gardens.

Phil Ramsden Dragons and Snowflakes Mathematics Department, Imperial College, London, UK Email: p.ramsden@imperial.ac.uk Bio: Phil Ramsden is a Content Consultant for London Gifted & Talented (which develops learning materials for London's brightest school students). Phil has been using Mathematica with students since Version 1.0 came out in the late '80s, and hosted the 2003 International Mathematica Symposium in London. He has experience in school and college teaching, and for the last eleven years has specialised in computer-based learning in Higher Education. He runs regular mathematical Summer Schools and on-site courses for high school students. Phil's mathematical specialisation is nonlinear dynamical systems and chaos.

Bart M. ter Haar Romeny Medical Image Analysis Biomedical Imaging, Eindhoven University of Technology Email: b.m.terhaarromeny@tue.nl Abstract: 80% of all diagnoses are done on medical images. There is a wide array of imaging modalities, such as Computer Tomography, Magnetic Resonance Imaging, Ultrasound, etc. The number of images is becoming overwhelming; several thousand CT slices can be generated in a total-body scan. New and effective algorithms are needed to assist the specialist in creating advanced interactive 3D visualizations (a kind of visual data compression) and to automatically analyze the images. Doctors have many questions about the images that can be answered by advanced computer vision algorithms. Computer-aided diagnosis and quantitative image analysis are emerging as new powerful tools, among others in mammography, lung cancer, polyp detection in the colon etc. Mathematica is an ideal tool for the design phase of such algorithms, due to their mathematical complexity. It has also been strikingly effective in teaching these techniques, due to the powerful graphical feedback and steep learning curve. We will discuss several medical applications, in a richly illustrated way. Bio: Bart M. ter Haar Romeny received an MSc degree in Applied Physics at Delft University, and a PhD degree (1983) at Utrecht University, the Netherlands. He served from 1983 till 2001 as principal physicist and associate professor at the Dept. of Radiology and the Image Sciences Institute at Utrecht University. He is now full professor at Eindhoven University of Technology, Dept. of Biomedical Engineering. His interests are medical image analysis and multi-scale computer vision, its mathematical foundations and clinical applications. A focus on the human visual system gives biological inspiration for the mathematics. He is the author of a popular interactive tutorial book on perceptually inspired multi-scale image analysis, written in Mathematica.

Eric Weisstein Making MathWorld Wolfram Research Inc Email: eww@wolfram.com Abstract: MathWorld is the internet's most popular mathematical resource. Its extensive and detailed entries usually find their way into the top 3 (if not top 1) results returned by a Google search for any given mathematical term. Mathematica has long been instrumental in the authoring of this website, which is hosted as a public service by Wolfram Research. As a result of new work, the website contents is now authored entirely in notebooks which are directly converted to web pages by Mathematica itself. In this talk, the processes that bring MathWorld from keyboard to the web will be discussed, focusing especially on the Mathematica-based tools that make this process possible. Bio: Eric Weisstein graduated from Cornell University (BA 1990) with a degree in physics and from the California Institute of Technology (MS 1993; PhD 1996) with degrees in planetary astronomy. Eric has been compiling scientific encyclopedias for more than 15 years. His mathematics encyclopedia — MathWorld — is part of a larger project to collect and make available detailed mathematical and scientific information in a way most accessible to lay people.

Tom Wickham-Jones Mathematica Development Tools Wolfram Research Inc Email: twj@wolfram.com Abstract: How to Boost the Productivity of the Mathematica Programmer Bio: Tom Wickham-Jones has worked for Wolfram Research on the implementation of Mathematica since 1990, he is currently the Director of Kernel Technology. He has worked on many details of the Mathematica programming language as well as other areas such as graphics. In addition he is the architect of webMathematica. In 1992 he published the book Mathematica Graphics: Techniques and Applications.