2020
Conference article  Restricted

Global explanations with local scoring

Setzu M., Guidotti R., Monreale A., Turini F.

Explainable AI  Rule-based explainer  Decision system 

Artificial Intelligence systems often adopt machine learning models encoding complex algorithms with potentially unknown behavior. As the application of these "black box" models grows, it is our responsibility to understand their inner working and formulate them in human-understandable explanations. To this end, we propose a rule-based model-agnostic explanation method that follows a local-to-global schema: it generalizes a global explanation summarizing the decision logic of a black box starting from the local explanations of single predicted instances. We define a scoring system based on a rule relevance score to extract global explanations from a set of local explanations in the form of decision rules. Experiments on several datasets and black boxes show the stability, and low complexity of the global explanations provided by the proposed solution in comparison with baselines and state-of-the-art global explainers.

Source: Joint European Conference on Machine Learning and Knowledge Discovery in Databases - ECML PKDD 2019, pp. 159–171, Würzburg, Germany, 16-20 September, 2019

Publisher: Springer, Heidelberg ;, Germania



Back to previous page
Projects (via OpenAIRE)

AI4EU
A European AI On Demand Platform and Ecosystem

Track and Know
Big Data for Mobility Tracking Knowledge Extraction in Urban Areas

PRO-RES
PROmoting integrity in the use of RESearch results

XAI
Science and technology for the explanation of AI decision making

SoBigData
SoBigData Research Infrastructure


OpenAIRE
BibTeX entry
@inproceedings{oai:it.cnr:prodotti:424675,
	title = {Global explanations with local scoring},
	author = {Setzu M. and Guidotti R. and Monreale A. and Turini F.},
	publisher = {Springer, Heidelberg ;, Germania},
	doi = {10.1007/978-3-030-43823-4_14},
	booktitle = {Joint European Conference on Machine Learning and Knowledge Discovery in Databases - ECML PKDD 2019, pp. 159–171, Würzburg, Germany, 16-20 September, 2019},
	year = {2020}
}