Reverse-Engineering Utilities from Preferences
Definition
The method of inferring an agent's utility function from a reliable subset of their observed preferences, then using it to derive recommendations for more complex decisions.
Cast within
No topic-region cast yet — this appears once Reverse-Engineering Utilities from Preferences's compiled claims are aligned into a topic region's argument tree.
Claims mentioning Reverse-Engineering Utilities from Preferences (2)
Your notes
Loading notes...