Skip to Main content Skip to Navigation
Conference papers

Explaining single predictions : a faster method

Abstract : Machine learning has proven increasingly essential in manyfields. Yet, a lot obstacles still hinder its use by non-experts. The lack oftrust in the results obtained is foremost among them, and has inspiredseveral explanatory approaches in the literature. In this paper, we areinvestigating the domain of single prediction explanation. This is per-formed by providing the user a detailed explanation of the attribute'sinfluence on each single predicted instance, related to a particular ma-chine learning model. A lot of possible explanation methods have beendeveloped recently. Although, these approaches often require an impor-tant computation time in order to be efficient. That is why we are inves-tigating about new proposals of explanation methods, aiming to increasetime performances, for a small loss in accuracy.
Document type :
Conference papers
Complete list of metadata

Cited literature [13 references]  Display  Hide  Download
Contributor : Open Archive Toulouse Archive Ouverte (oatao) Connect in order to contact the contributor
Submitted on : Thursday, September 17, 2020 - 5:00:56 PM
Last modification on : Wednesday, November 3, 2021 - 6:50:56 AM
Long-term archiving on: : Thursday, December 3, 2020 - 10:24:21 AM


Files produced by the author(s)



Gabriel Ferrettini, Julien Aligon, Chantal Soulé-Dupuy. Explaining single predictions : a faster method. International Conference on Current Trends in Theory and Practice of Computer Science (SOFSEM 2020), Jan 2020, Limassol, Cyprus. pp.313-324, ⟨10.1007/978-3-030-38919-2_26⟩. ⟨hal-02942310⟩



Record views


Files downloads