Some Notes on Ethics 2024-02-01T10:18:44.502Z

A New Bayesian Decision Theory 2023-09-20T09:36:50.771Z

Decision Theory: A (Normative) Introduction 2023-09-06T08:22:57.877Z

A Defense of Utilitarianism 2023-04-06T21:09:59.927Z

Comment by Pareto Optimal on Decision Theory: A (Normative) Introduction · 2023-09-06T08:29:34.195Z · LW · GW

Broadly speaking, without regard to labels or choosing a specific algorithm, I think one should choose the dominance principle if probabilities are held constant and thus choose smoking, two-boxing, etc. But if probabilities were to update based on the agent's actions such as in Parfit's Hitchhiker, then one should pre-commit to an action that does not necessarily dominate in exchange for an outcome that is (second?) best and thus pay the biker. 

We can of course play semantic slapfights and call this something like 'Bayesian Decision Theory' but ideas matter more than labels.