User: Guest  Login
Title:

When are post-hoc conceptual explanations identifiable?

Document type:
Konferenzbeitrag
Author(s):
Leemann, Tobias; Kirchhof, Michael; Rong, Yao; Kasneci, Enkelejda; Kasneci, Gjergji
Abstract:
Interest in understanding and factorizing learned embedding spaces through conceptual explanations is steadily growing. When no human concept labels are available, concept discovery methods search trained embedding spaces for interpretable concepts like object shape or color that can provide post-hoc explanations for decisions. Unlike previous work, we argue that concept discovery should be identifiable, meaning that a number of known concepts can be provably recovered to guarantee reliability o...     »
Editor:
Evans, Robin J.; Shpitser, Ilya
Book / Congress title:
Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence
Volume:
216
Publisher:
PMLR
Year:
2023
Month:
31 Jul--04 Aug
Pages:
1207--1218
Bookseries title:
Proceedings of Machine Learning Research
WWW:
https://proceedings.mlr.press/v216/leemann23a.html
 BibTeX