Using prototypes to improve convolutional networks interpretability

Thalita Drumond 1, 2, 3 Thierry Viéville 1, 2, 3 Frédéric Alexandre 1, 2, 3
1 Mnemosyne - Mnemonic Synergy
LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest, IMN - Institut des Maladies Neurodégénératives [Bordeaux]
Abstract : We propose a method that allows the interpretation of the data representation obtained by CNN, through introducing prototypes in the feature space, that are later classified into a certain category. This way we can see how the feature space is structured in link with the categories and the related task.
Document type :
Conference papers
Complete list of metadatas

Cited literature [23 references]  Display  Hide  Download

https://hal.inria.fr/hal-01651964
Contributor : Thierry Viéville <>
Submitted on : Wednesday, November 29, 2017 - 5:32:58 PM
Last modification on : Friday, May 25, 2018 - 2:59:26 PM

File

main_NIPS_2017.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01651964, version 1

Citation

Thalita Drumond, Thierry Viéville, Frédéric Alexandre. Using prototypes to improve convolutional networks interpretability. NIPS 2017 - 31st Annual Conference on Neural Information Processing Systems: Transparent and interpretable machine learning in safety critical environments Workshop, Dec 2017, Long Beach, United States. ⟨hal-01651964⟩

Share

Metrics

Record views

363

Files downloads

405