[SEQ RERUN] My Bayesian Enlightenment — LessWrong