Why did you like it better than other courses? In my case, I really liked stochastic geometry. It was quite theoretical, e.g. estimating the value of Pi by throwing a needle in a rectangular grid and counting how many times the needle hit one of the lines in the grid. But it had some nice applications too. One of the problems (known as the third Grenander Problem or Exterior/Interior Problem) consisted of estimating the shape of an oil field: about 20 oil wells had been digged, and we knew which ones had oil, which ones didn't. The only assumption was that the shape of the oil field was convex. We eventually developed a fast algorithm to compute a convex domain in any dimension (when dimension = 2 or 3, the problem is easy).

As you can imagine, the issue was trying to estimate the shape / extent of oil field with as few wells as possible, as digging wells outside the field is expensive.