A small noise can have a dramatic effect, overpowering deterministic dynamics and leading to so-called rare events.  We will recall some results from the Freidlin-Wentzell theory of large deviations, and describe some recent results, drawing connections with applied mathematics and the calculus of variations.