The next C3.ai Digital Transformation Institute Colloquia on Digital Transformation Science will be Thursday, February 18 at 3:00 p.m. U.S. Central time. Presenting "Why Do ML Models Fail?" will be Aleksander Madry from the Massachusetts Institute of Technology. Registration is required to attend this event.
Abstract: Our current machine learning (ML) models achieve impressive performance on many benchmark tasks. Yet, these models remain remarkably brittle, susceptible to manipulation and, more broadly, often behave in ways that are unpredictable to users. Why is this the case? In this talk, we identify human-ML misalignment as a chief cause of this behavior. We then take an end-to-end look at the current ML training paradigm and pinpoint some of the roots of this misalignment. We discuss how current pipelines for dataset creation, model training, and system evaluation give rise to unintuitive behavior and widespread vulnerability. Finally, we conclude by outlining possible approaches towards alleviating these deficiencies.
