Distinguished Lecture: Some excursions into interpretable machine learning by Sanjoy Dasgupta, UC San Diego
The need for interpretability in machine learning opens up a host of
new algorithmic and statistical challenges. We will see how these emerge
on three fronts: (i) interpretable clustering, (ii) interpretable
classification with logical rules, and (iii) statistical estimation
of the quality of black-box explanation systems.
Sanjoy Dasgupta is Professor of Computer Science at UC San Diego. He
works primarily on unsupervised and minimally supervised learning.
He is the author of a textbook, Algorithms, with Christos Papadimitriou
and Umesh Vazirani.