Premium Partner

Session Information

Interpretable Deep Learning

Speaker(s):

Dr. Wojciech Samek (Fraunhofer Heinrich Hertz Institute)
Back to last page:
 
Deep neural networks (DNNs) have demonstrated high predictive performance on a number of tasks in the sciences and industry. However, these predictive models often behave as black-boxes, i.e., it is hard to grasp what makes them arrive at a particular decision. On practical problems where a single incorrect prediction can be costly, a simple prediction cannot be trusted by default. Instead, the prediction should be made interpretable to a human expert for careful verification.This talk presents a recently developed technique which allows to visualize and interpret the result of DNN inference. Different applications of this method to computer vision are presented.
Audience adressed: 
TBA
Takeaway: 

TBA

Session Type: 
Talk
Track: 
Code / Platforms / Tools
Experience level adressed: 
Intermediate
Time slot: 
2016-04-19 16:00
Room: 
Channel 5

Event Sponsors

Sponsor Partners

No sponsors in this category have been announced yet.

Media and Partners

Edit User
<