Reasoning and Planning with Dynamically Changing Norms
This work tackles the underexplored problem of norm-guided planning with dynamic norms for human-AI interaction, but the empirical evaluation is limited to a single dialogue task.
The paper addresses the problem of guiding AI planning with dynamically changing norms in human-AI settings, proposing a defeasible calculus for normative conflict resolution and using norms as guard rails. The approach is demonstrated theoretically with formal proofs and empirically with the SocialBot agent on a natural language dialogue task.
To safely interact with humans, AI agents must both know our norms and consider them during planning. However, such norm-guided planning has been less explored, only within communities of artificial agents, and has ignored the dynamic nature of norms. This paper instead presents an approach to guiding planning with dynamically changing norms in a human-AI setting. We contribute a defeasible calculus for resolving normative conflicts and an approach to using such dynamically changing norms as guard rails on plans. We theoretically demonstrate our approach with formal proofs and empirically with an AI agent, SocialBot, on a natural language dialogue task.