Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Gianluigi Lopardo

Gianluigi Lopardo

Understanding Post-hoc Explainers: The Case of Anchors

Mar 15, 2023
Gianluigi Lopardo, Frederic Precioso, Damien Garreau

Figure 1 for Understanding Post-hoc Explainers: The Case of Anchors

Figure 2 for Understanding Post-hoc Explainers: The Case of Anchors

* arXiv admin note: substantial text overlap with arXiv:2205.13789

Access Paper or Ask Questions

Comparing Feature Importance and Rule Extraction for Interpretability on Text Data

Jul 04, 2022
Gianluigi Lopardo, Damien Garreau

Figure 1 for Comparing Feature Importance and Rule Extraction for Interpretability on Text Data

Figure 2 for Comparing Feature Importance and Rule Extraction for Interpretability on Text Data

Figure 3 for Comparing Feature Importance and Rule Extraction for Interpretability on Text Data

Figure 4 for Comparing Feature Importance and Rule Extraction for Interpretability on Text Data

* Accepted to XAIE ICPR 2022, the 2-nd Workshop on Explainable and Ethical AI, ICPR 2022

Access Paper or Ask Questions

A Sea of Words: An In-Depth Analysis of Anchors for Text Data

May 27, 2022
Gianluigi Lopardo, Damien Garreau, Frederic Precioso

Figure 1 for A Sea of Words: An In-Depth Analysis of Anchors for Text Data

Figure 2 for A Sea of Words: An In-Depth Analysis of Anchors for Text Data

Figure 3 for A Sea of Words: An In-Depth Analysis of Anchors for Text Data

Figure 4 for A Sea of Words: An In-Depth Analysis of Anchors for Text Data

* 10+2 page paper, 15-page appendix

Access Paper or Ask Questions

SMACE: A New Method for the Interpretability of Composite Decision Systems

Nov 16, 2021
Gianluigi Lopardo, Damien Garreau, Frederic Precioso, Greger Ottosson

Figure 1 for SMACE: A New Method for the Interpretability of Composite Decision Systems

Figure 2 for SMACE: A New Method for the Interpretability of Composite Decision Systems

Figure 3 for SMACE: A New Method for the Interpretability of Composite Decision Systems

Figure 4 for SMACE: A New Method for the Interpretability of Composite Decision Systems

* 7 pages, 3 figures

Access Paper or Ask Questions