Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
Add to list
  1. Search results
  2. The need for a numeric measure of exp...

The need for a numeric measure of explainability

Wim De Mulder and Peggy Valcke
(2021) 2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA). In IEEE International Conference on Big Data p.2712-2720
Author
Organization
Abstract
The attempt to concretely define the concept of explainability in terms of other vaguely described notions, is doomed to fail. We argue that a rigorous numeric measure is needed to take a significant step forward. Such a measure would allow the objective comparison of machine learning models that are explainable to varying degrees. In the longer term, a numeric measure might make it possible to compare humans and machines in terms of explanatory capacities, thereby establishing a very specific kind of Turing test. We introduce some notions and rigorous definitions related to explainability, in particular completeness and consistency. The objective of this paper is that the introduced terms can act as a concrete starting point for the construction of a mathematically expressed measure of explainability. We also show how the idea of model selection, where a model is chosen from a candidate set by optimizing the trade-off between complexity and goodness of fit, can be applied to select explainable models.
Keywords
explainability, completeness, consistency, model selection, AUTOMATED DECISION-MAKING, MODEL SELECTION, BLACK-BOX, EXPLANATION, ALGORITHM, ENTROPY, IMPACT, CANCER

Citation

Please use this url to cite or link to this publication:

MLA
De Mulder, Wim, and Peggy Valcke. “The Need for a Numeric Measure of Explainability.” 2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), IEEE, 2021, pp. 2712–20, doi:10.1109/BigData52589.2021.9671745.
APA
De Mulder, W., & Valcke, P. (2021). The need for a numeric measure of explainability. 2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2712–2720. https://doi.org/10.1109/BigData52589.2021.9671745
Chicago author-date
De Mulder, Wim, and Peggy Valcke. 2021. “The Need for a Numeric Measure of Explainability.” In 2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2712–20. IEEE. https://doi.org/10.1109/BigData52589.2021.9671745.
Chicago author-date (all authors)
De Mulder, Wim, and Peggy Valcke. 2021. “The Need for a Numeric Measure of Explainability.” In 2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2712–2720. IEEE. doi:10.1109/BigData52589.2021.9671745.
Vancouver
1.
De Mulder W, Valcke P. The need for a numeric measure of explainability. In: 2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA). IEEE; 2021. p. 2712–20.
IEEE
[1]
W. De Mulder and P. Valcke, “The need for a numeric measure of explainability,” in 2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), Online, 2021, pp. 2712–2720.
@inproceedings{01GMGJQNJ129DYKKT97V80YQ5D,
  abstract     = {{The attempt to concretely define the concept of explainability in terms of other vaguely described notions, is doomed to fail. We argue that a rigorous numeric measure is needed to take a significant step forward. Such a measure would allow the objective comparison of machine learning models that are explainable to varying degrees. In the longer term, a numeric measure might make it possible to compare humans and machines in terms of explanatory capacities, thereby establishing a very specific kind of Turing test.
We introduce some notions and rigorous definitions related to explainability, in particular completeness and consistency. The objective of this paper is that the introduced terms can act as a concrete starting point for the construction of a mathematically expressed measure of explainability. We also show how the idea of model selection, where a model is chosen from a candidate set by optimizing the trade-off between complexity and goodness of fit, can be applied to select explainable models.}},
  author       = {{De Mulder, Wim and Valcke, Peggy}},
  booktitle    = {{2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)}},
  isbn         = {{9781665445993}},
  issn         = {{2639-1589}},
  keywords     = {{explainability,completeness,consistency,model selection,AUTOMATED DECISION-MAKING,MODEL SELECTION,BLACK-BOX,EXPLANATION,ALGORITHM,ENTROPY,IMPACT,CANCER}},
  language     = {{eng}},
  location     = {{Online}},
  pages        = {{2712--2720}},
  publisher    = {{IEEE}},
  title        = {{The need for a numeric measure of explainability}},
  url          = {{http://doi.org/10.1109/BigData52589.2021.9671745}},
  year         = {{2021}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: