Finale Doshi-Velez

A Roadmap for the Rigorous Science of Interpretability

Finale Doshi-Velez

Talk: A Roadmap for the Rigorous Science of Interpretability

With a growing interest in interpretability, there is an increasing need to characterize what exactly we mean by it and how to sensibly compare the interpretability of different approaches. In this talk, Finale will start by discussing some research in interpretable machine learning from her group, and then broaden it out to discuss what interpretability is and when it is needed. She'll argue that the desire for "interpretability" is as vague as asking for "good predictions". This objective of this talk is to start a conversation to do the same for interpretability: she will suggest a taxonomy for interpretable models and their evaluation, and also highlight important open questions about the science of interpretability in machine learning.

Finale Doshi-Velez

Finale Doshi-Velez is an assistant professor in Computer Science at Harvard University. She completed her postdoc at the Harvard Medical Center.