Mihaela van der Schaar and postdoc Fergus Imrie will deliver a tutorial at the Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI-22). Additionally, Mihaela van der Schaar will give an invited talk at the Workshop on Information-Theoretic Methods for Causal Inference and Discovery (ITCI’22), and one of the lab’s recent papers has been accepted for publication at AAAI-22.
ITCI’22 invited talk
One of the lab’s recent papers has been accepted for publication at AAAI-22; details are provided below.
Inferring Lexicographically-Ordered Rewards from Preferences
Alihan Hüyük, William R. Zame, Mihaela van der Schaar
Modeling the preferences of agents over a set of alternatives is a principal concern in many areas. The dominant approach has been to find a single reward/utility function with the property that alternatives yielding higher rewards are preferred over alternatives yielding lower rewards. However, in many settings, preferences are based on multiple—often competing—objectives; a single reward function is not adequate to represent such preferences.
This paper proposes a method for inferring multi-objective reward-based representations of an agent’s observed preferences. We model the agent’s priorities over different objectives as entering lexicographically, so that objectives with lower priorities matter only when the agent is indifferent with respect to objectives with higher priorities.
We offer two example applications in healthcare—one inspired by cancer treatment, the other inspired by organ transplantation—to illustrate how the lexicographically-ordered rewards we learn can provide a better understanding of a decision-maker’s preferences and help improve policies when used in reinforcement learning.
The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. AAAI-22 will have a diverse technical track, student abstracts, poster sessions, invited speakers, tutorials, workshops, and exhibit and competition programs, all selected according to the highest reviewing standards. AAAI-22 welcomes submissions on mainstream AI topics as well as novel crosscutting work in related areas.
For a full list of the van der Schaar Lab’s publications, click here.