Towards a Principled Understanding of Deep Learning

Towards a Principled Understanding of Deep Learning

Internship Description

Deep learning models provide state of the art performance on many practical machine learning tasks. However, there is a large gap between our theoretical / conceptual understanding and practice.

The intern will work in one of the follow areas, depending on interest and background:

 -  deep learning models

-  adversarial attacks and robustness

-  optimization for deep learning

-  generalization of deep learning

-  GANs

-  model compression

Deliverables/Expectations

Ideally contribution to a research paper​.

Faculty Name

Peter Richtarik

Field of Study

computer science, mathematics