Texas A & M

A very common problem in Science is that we have some Data Observations and we are interested in either approximating the function underlying the data or computing some quantity of interest about this function.  This talk will discuss what are best algorithms for such tasks and how we can evaluate the performance of any such algorithm.