Context-specific value inference via hybrid intelligence

Abstract

Human values are the abstract motivations that drive our opinions and actions. AI agents ought to align their behavior with our value preferences (the relative importance we ascribe to different values) to co-exist with us in our society. However, value preferences differ across individuals and are dependent on context. To reflect diversity in society and to align with contextual value preferences, AI agents must be able to discern the value preferences of the relevant individuals by interacting with them. We refer to this as the value inference challenge, which is the focus of this thesis. Value inference entails several challenges and the related work on value inference is scattered across different AI subfields. We present a comprehensive overview of the value inference challenge by breaking it down into three distinct steps and showing the interconnections among these steps.