Towards Understanding the Inner Workings of Deep Neural Networks

2:10 pm - 2:30 pm

Objective: Guidance
Audience Level: Beginner/Intermediate
Session Type: Presentation

Artificial Neural Networks are a powerful class of models with advantages for many problems of interest. I will discuss research that sheds light on the inner workings of neural nets, and shows how can we make sense of non-linear computations performed by millions of parameters. These studies reveal surprising features of neural nets and lead to a greater understanding of and intuitions of them.


, PhD candidate, Cornell University