Learn · CERN

Talk24 min

Identifying the relevant dependencies of the neural network response on characteristics of the input space

Stefan Wunsch · CERN

Watch At CERN

About

The use of neural networks in physics analyses poses new challenges for the estimation of systematic uncertainties. Since the key to a proper estimation of uncertainties is the precise understanding of the algorithm, novel methods for the detailed study of the trained neural network are valuable. This talk presents an approach to identify those characteristics of the neural network inputs that are most relevant for the response and therefore provides essential information to determine the systematic uncertainties..

0:00:00 Slide 1

0:00:38 Slide 2

0:01:49 Slide 3

0:02:37 Slide 4

0:04:46 Slide 5

0:06:08 Slide 6

0:06:15 Slide 7

0:06:20 Slide 8

0:07:12 Slide 9

0:07:29 Slide 10

0:07:35 Slide 11

The talk

  1. Identifying the relevant dependencies of the neural network response on characteristics of the input space 2018-04-10 · 24 min · CERN's player

Credits and licence

Institution
CERN
Department
Machine Learning
Speaker
Stefan Wunsch
Year
2018
Language
English
Subjects
Computer science
Source
CERN
Licence
CERN, as CERN states it. CERN's terms
Attribution
© CERN. CERN provides its audiovisual media free of charge for educational and informational use, credited to CERN; this lecture is shown in CERN's own player.
Who can watch
CERN gives no rating. Invisible Network shows it to Tweens profiles and older, judged from what the source says about it. How courses are placed