SILO: Towards Plurality: Foundations for Learning from Diverse Human Preferences

Large pre-trained models trained on internet-scale data are often not ready for safe deployment out of the box. They are heavily fine-tuned and aligned using large quantities of human preference data. When we want to align an AI/ML model to human preference or values,  it is worthwhile to ask whose preference and values we are aligning it to? Recently, the limitations of current approaches due to their inherent uniformity assumption have been highlighted and the need for plurality – capturing the diversity in human preferences and values – is getting recognized as an important challenge to address. While alignment from human preferences has currently become a very active area of research, it is worthwhile to note that there is rich literature on learning preferences from human judgements using comparison queries. It plays a crucial role in several applications ranging from cognitive and behavioral psychology, crowdsourcing democracy, surveys in social science applications, and recommendation systems. However, the models in the literature often focus on learning average preference over the population due to the limitations on the amount of data available per individual and also assume the knowledge of the metric, i.e., the way humans judge similarity and dissimilarity.

In this talk, I will discuss some recent results that focus on how we can reliably capture diversity in preferences while pooling together data from individuals. In particular, I will talk about fundamental questions in two directions: (1) Simultaneous metric and preference learning where the goal is to learn an unknown but shared metric from preference queries while the preferences are diverse and also unknown. (2) Learning distribution of preferences over a population with a single comparison query per individual.

Bio:

Ramya Korlakai Vinayak is an assistant professor in the Dept. of ECE and affiliated faculty in the Dept. of Computer Science and the Dept. of Statistics at the UW-Madison. Her research interests span the areas of machine learning, statistical inference, and crowdsourcing. Her work focuses on addressing theoretical and practical challenges that arise when learning from societal data. Prior to joining UW-Madison, Ramya was a postdoctoral researcher in the Paul G. Allen School of Computer Science and Engineering at the University of Washington. She received her Ph.D. in Electrical Engineering from Caltech. She obtained her Masters from Caltech and Bachelors from IIT Madras. She is a recipient of the Schlumberger Foundation Faculty of the Future fellowship from 2013-15, and an invited participant at the Rising Stars in EECS workshop in 2019. She is the recipient of NSF CAREER Award 2023-2028.