Events and Seminars

We are thrilled to announce that CAIS will host Richard Zemel, the Trianthe Dakolias Professor of Engineering and Applied Science in the Computer Science Department at Columbia University. He is the Director of the NSF AI Institute for Artificial and Natural Intelligence (ARNI). He was the Co-Founder and inaugural Research Director of the Vector Institute for Artificial Intelligence. He is a Canadian Institute for Advanced Research AI Chair, an Amazon Scholar, and is on the Advisory Board of the Neural Information Processing Society.  His research contributions include foundational work on systems that learn useful representations of data with little or no supervision; graph-based machine learning; and algorithms for fair and robust machine learning.

TITLE:

“A Framework for Responsible Deployment of AI Models”

ABSTRACT:

Explicit statistical guarantees on model performance are an important ingredient in responsible machine learning. I will describe a simple and flexible framework that allows us to construct rigorous bounds on the loss of deployed prediction models.  We have developed the framework with the aim of applying it to a number of areas.  One area is algorithmic fairness, where it can control the extent to which different members of a population experience unequal effects of algorithmic decisions.  A second application area is pretrained language models, where the framework helps determine how best to prompt a model to perform a given task. Experiments on applications such as chatbots, medical question summarization, and code generation highlight how such a framework can reduce the risk of the worst outcomes. Finally I will discuss how generative AI models can be tested with respect to fairness, and how they relate to current anti-discrimination regulation.

When: Wednesday, November 20th, 2024
Where: York University, Keele Campus, Osgoode Hall Law School, Room 2027

Time: 11:00 AM – 12:30 PM