The Bayes Theorem, expressed as P(O|E) = P(E|O)P(O)/P(E), describes a relationship between probabilities and conditional probabilities of two events. In Machine Learning, it is often used to compute the probability of a certain outcome (the so-called 'posterior probability' -- P(O|E) in the equation) given newly observed 'evidence', based on: (1) the 'prior' probability of the outcome, before the evidence was observed -- P(O) in the equation; (2) the 'likelihood' that the evidence would be observed assuming the outcome holds -- P(E|O) in the equation; (3) the probability of the evidence, considering all possible outcomes -- P(E) in the equation.