Events

In this thesis, we revisit the task of data selection through the lens of social-oriented metrics like diversity and fairness.

In this dissertation, I explore how lightweight dynamic analysis can be used to improve the reliability of Python software.

In the first part of this thesis, we formulate a general-purpose information-theoretic criterion that allows further improving SSL representations.

A historical overview of poisoning attacks that manipulate training data to compromise the performance of machine learning systems at test time, along with...

Meetings will occur from 12:00 p.m.–1:30 p.m. on Mondays; lunch will be provided and will be available starting at 11:45 a.m.