10:30 AM-12:30 PM
Room: Atlanta 3
In many scientific computing problems, it is desirable to get guaranteed estimates. For such problems, researchers have developed computational methods with automatic results verification (also known as "interval computations"). These methods have been successfully applied to many practical problems, but at present, they are not yet so widely applied as more traditional computational techniques. One of the possible ways to drastically increase the scope of verified methods is to combine them with more traditional techniques. In this minisymposium, the speakers describe how interval computations can be combined with traditional numerical techniques, with (probability-motivated) maximum entropy methods, and with geometric computations.
Organizers: Vladik Kreinovich and Scott A. Starks
University of Texas, El Paso
MMD, 3/26/99
tjf, 1/19/99