15 ms·
I think it's probably reasonable to say it's an axiom. That's compatible with my claim that it's an arbitrary choice that is not inherently more correct nor bet
by rcthompson 9y ago
I think it's probably reasonable to say it's an axiom. That's compatible with my claim that it's an arbitrary choice that is not inherently more correct nor better supported by any information than any other choice, except perhaps in the sense that it feels better or more intuitive.
- skissane 9y agoWell, what happens if we adopt alternative axioms (e.g. some sort of assumption of a normal distribution instead of a uniform one)? Are those alternative axioms consistent? Maybe we might be able to show that agents adopting the uniform axiom make better (more rewarding) decisions in practice than those who adopt a competing axiom do? I don't have a proof, but my gut says yes.