Cancelled: Talk: Eyke Hüllermeier (LMU Munich)

Title:

Preference, Decision, and Choice: A Machine Learning Perspective

Abstract:

The topic of "preferences" has attracted increasing attention in machine learning (and AI in general) in the recent past, where preference learning has emerged as an interdisciplinary research field with close connections to operations research, social choice, and the decision sciences. Roughly speaking, preference learning is about methods for learning preference models from explicit or implicit information about the preferences of an individual or a group of individuals, and the use of such models for decision-making and prediction, amongst others. I will start this talk with an introduction to preference learning and a brief overview of the field. Two recent developments and topics of ongoing work will then be discussed in more detail, namely, the learning of context-dependent preferences (where the preference for a choice alternative may depend on the availability of other alternatives) and the learning of choice functions via a Pareto-embedding of choice alternatives.