x
Two Notions of Best Response — LessWrong