English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
 Local TagsRelease HistoryDetailsSummary
  Lack of Robustness in Artificial Neural Networks

Bethge, M. (2019). Lack of Robustness in Artificial Neural Networks. Neuroforum, 25(Supplement 1): S23-1, 179.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://hdl.handle.net/21.11116/0000-0003-1F71-C Version Permalink: https://hdl.handle.net/21.11116/0000-0004-AA6D-3
Genre: Meeting Abstract

Files

show Files

Locators

show
hide
Locator:
https://www.nwg-goettingen.de/2019/upload/file/Proceedings_NWG2019.pdf (Abstract)
Description:
-
OA-Status:

Creators

show
hide
 Creators:
Bethge, M1, Author           
Affiliations:
1External Organizations, ou_persistent22              

Content

show
hide
Free keywords: -
 Abstract: Deep neural networks have become a ubiquitous tool in a broad range of AI applications. Resembling important aspects of rapid feed-forward visual processing in the ventral stream they can be trained to
match human behavior on standardized pattern recognition tasks. Outside the training distribution, however, decision making of artificial neural networks exhibits large discrepancies to biological vision
systems. I will give an overview on the lack of robustness in deep neural networks and present recent results of my lab to quantify and overcome these discrepancies.

Details

show
hide
Language(s):
 Dates: 2019-02
 Publication Status: Issued
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: -
 Degree: -

Event

show
hide
Title: 13th Göttingen Meeting of the German Neuroscience Society, 37th Göttingen Neurobiology Conference
Place of Event: Göttingen, Germany
Start-/End Date: 2019-03-20 - 2019-03-23

Legal Case

show

Project information

show

Source 1

show
hide
Title: Neuroforum
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: -
Pages: - Volume / Issue: 25 (Supplement 1) Sequence Number: S23-1 Start / End Page: 179 Identifier: ISSN: 0947-0875
CoNE: https://pure.mpg.de/cone/journals/resource/110978984249776