Research news on Monte Carlo methods

Monte Carlo methods are a class of computational techniques that use random sampling to approximate solutions to mathematical and statistical problems that are analytically intractable. They typically involve defining a probabilistic model of the system or quantity of interest, drawing large numbers of independent samples from relevant distributions (often via pseudorandom number generators and sampling algorithms such as rejection sampling or Markov chain Monte Carlo), and estimating expectations, integrals, probabilities, or optimization objectives from the empirical distribution of outcomes. Accuracy improves with the number of samples, with convergence properties characterized by laws of large numbers and central limit theorems.

Monte Carlo simulations bring new focus to electron microscopy

With highly specialized instruments, we can see materials on the nanoscale—but we can't see what many of them do. That limits researchers' ability to develop new therapeutics and new technologies that take advantage of their ...