In this talk I will give proofs of some 'invariance principles' in probability; the Central Limit Theorem, the Berry--Esseen Theorem, and multidimensional and higher-degree versions thereof. I will discuss these proofs from a computer science perspective, and show some applications to fields such as property testing, derandomization, learning, and inapproximability.