 It's very important to be able to estimate, not just produce the best guess for what's going on, but say, you know, this guess is probably this correct. Most of the statistical methods that we use on a day-to-day basis were developed before the advent of modern computing. And in order to make the necessary calculations feasible, these statistical methods assume this simple bell-curve-shaped uncertainty, and this allows you to do a lot of mathematics with pen and paper that you couldn't do otherwise. But now that we have, you know, high-performance computers at our fingertips, we don't need to rely on these approximations anymore. And there are some problems in which those approximations can actually be catastrophically bad. And density estimation, the problem we address in our paper, is one of those problems. So what we did is using tools from theoretical physics, an area called field theory. We developed a method and implemented it in software that allows us to exactly quantify not only the best estimate for probability distribution from a small amount of data, but how uncertain we are in that estimate. And that's really the heart of statistics in science, is not just being able to make claims, but being able to state how confident you are in your claims.