dkirmani comments on Why Do I Think I Have Values?

[dkirmani]

What would “incoherent decisions” look like for an agent that has a utility function defined on action-observation histories?