Bayesian Buddhism: a path to optimal enlightenment — LessWrong