Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
GAN-Based Generation and Automatic Selection of Explanations for Neural Networks
Queen Mary University of London.
Queen Mary University of London.
Queen Mary University of London.
KTH, School of Electrical Engineering and Computer Science (EECS), Speech, Music and Hearing, TMH.
Show others and affiliations
2019 (English)Conference paper, Published paper (Refereed)
Abstract [en]

One way to interpret trained deep neural networks (DNNs) is by inspecting characteristics that neurons in the model respond to, such as by iteratively optimising themodelinput(e.g.,animage)tomaximallyactivatespecificneurons. However, this requires a careful selection of hyper-parameters to generate interpretable examples for each neuron of interest, and current methods rely on a manual, qualitative evaluation of each setting, which is prohibitively slow. We introduce a new metricthatusesFr´echetInceptionDistance(FID)toencouragesimilaritybetween model activations for real and generated data. This provides an efficient way to evaluateasetofgeneratedexamplesforeachsettingofhyper-parameters. Wealso propose a novel GAN-based method for generating explanations that enables an efficient search through the input space and imposes a strong prior favouring realistic outputs. We apply our approach to a classification model trained to predict whether a music audio recording contains singing voice. Our results suggest that thisproposedmetricsuccessfullyselectshyper-parametersleadingtointerpretable examples, avoiding the need for manual evaluation. Moreover, we see that examples synthesised to maximise or minimise the predicted probability of singing voice presence exhibit vocal or non-vocal characteristics, respectively, suggesting that our approach is able to generate suitable explanations for understanding concepts learned by a neural network.

Place, publisher, year, edition, pages
2019.
Keywords [en]
machine learning, interpretability, explainable AI
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-256496OAI: oai:DiVA.org:kth-256496DiVA, id: diva2:1346000
Conference
Safe Machine Learning 2019 Workshop at the International Conference on Learning Representations
Note

QC 20190925

Available from: 2019-08-26 Created: 2019-08-26 Last updated: 2019-09-25Bibliographically approved

Open Access in DiVA

fulltext(1709 kB)2 downloads
File information
File name FULLTEXT01.pdfFile size 1709 kBChecksum SHA-512
9ef680c56b92532c0b03c83101c3cdd60c9ed64a67ff6a2a17de6b1244379b099bb6e68eddcb4ee3e3a0dfb0191d9ae74633e82b645564bff3da14778cfa17d0
Type fulltextMimetype application/pdf

Other links

https://sites.google.com/view/safeml-iclr2019/accepted-papers

Search in DiVA

By author/editor
Sturm, Bob
By organisation
Speech, Music and Hearing, TMH
Other Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 2 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 26 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf