Mostrar el registro sencillo del ítem

LRP-Based path relevances for global explanation of deep architectures

dc.contributor.authorGuerrero-Gómez-Olmedo R.
dc.contributor.authorSalmeron J.L.
dc.contributor.authorKuchkovsky C.
dc.date.accessioned2020-09-02T22:19:31Z
dc.date.available2020-09-02T22:19:31Z
dc.date.issued2020
dc.identifier10.1016/j.neucom.2019.11.059
dc.identifier.citation381, , 252-260
dc.identifier.issn09252312
dc.identifier.urihttps://hdl.handle.net/20.500.12728/4764
dc.descriptionUnderstanding what Machine Learning models are doing is not always trivial. This is especially true for complex models such as Deep Neural Networks (DNN), which are the best-suited algorithms for modeling very complex and nonlinear relationships. But this need to understand has become a must since privacy regulations are hardening the industrial use of these models. There are different techniques to address the interpretability issues that Machine Learning models arises. This paper is focused on opening the so-called Deep Neural architectures black-box. This research extends the technique called Layer-wise Relevant Propagation (LRP) enhancing its properties to compute the most critical paths in different deep neural architectures using multicriteria analysis. We call this technique Ranked-LRP and it was tested on four different datasets and tasks, including classification and regression. The results show the worth of our proposal. © 2019
dc.language.isoen
dc.publisherElsevier B.V.
dc.subjectDeep learning
dc.subjectExplainable AI
dc.subjectInterpretable machine learning
dc.subjectLayer-wise relevant propagation
dc.subjectBackpropagation
dc.subjectClassification (of information)
dc.subjectComplex networks
dc.subjectDeep learning
dc.subjectMachine learning
dc.subjectNetwork architecture
dc.subjectDeep architectures
dc.subjectInterpretability
dc.subjectLayer-wise
dc.subjectMachine learning models
dc.subjectMulti Criteria Analysis
dc.subjectNeural architectures
dc.subjectNon-linear relationships
dc.subjectPrivacy regulation
dc.subjectDeep neural networks
dc.subjectarticle
dc.subjectdeep learning
dc.titleLRP-Based path relevances for global explanation of deep architectures
dc.typeArticle


Ficheros en el ítem

Thumbnail
Nombre:
item_2-s2.0-85076574056.pdf
Tamaño:
2.581Kb
Formato:
PDF
Ver/

Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem