Machine Learning in Montpellier, Theory & Practice – Nicolas Papernot (Univ. Toronto)
The talk first illustrates the challenges of having end users trust that machine learning algorithms were deployed responsibly, i.e., in a trustworthy way, through a deep dive on the problem of unlearning. The need for machine unlearning, i.e., obtaining a model one would get without training on a subset of data, arises from privacy legislation and more recently as a potential solution to data poisoning or copyright claims. As we present different approaches to unlearning, it becomes clear that they fail to answer our motivating question: how can end users verify that unlearning was successful? Taking a step back, we draw lessons for the broader area of trustworthy machine learning and present ongoing research that lay the foundations for companies, regulators, and countries to be able to verify meaningful properties at the scale that is required for stable governance of AI algorithms, both nationally and internationally.
