No silver bullet: interpretable ML models must be explained

Autor: Joao Marques-Silva, Alexey Ignatiev

Publikováno v: Frontiers in Artificial Intelligence, Vol 6 (2023)

Recent years witnessed a number of proposals for the use of the so-called interpretable models in specific application domains. These include high-risk, but also safety-critical domains. In contrast, other works reported some pitfalls of machine lear

Externí odkaz: https://doaj.org/article/fffc6a841f514a7f80af38a4fbcdd76f

Using MaxSAT for Efficient Explanations of Tree Ensembles

Autor: Alexey Ignatiev, Yacine Izza, Peter J. Stuckey, Joao Marques-Silva

Publikováno v: Proceedings of the AAAI Conference on Artificial Intelligence. 36:3776-3785

Tree ensembles (TEs) denote a prevalent machine learning model that do not offer guarantees of interpretability, that represent a challenge from the perspective of explainable artificial intelligence. Besides model agnostic approaches, recent work pr

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::44c2157d09a0298c97c40169f1c889f6
https://doi.org/10.1609/aaai.v36i4.20292

Learning Optimal Decision Sets and Lists with SAT

Autor: Jinqiang Yu, Alexey Ignatiev, Peter J. Stuckey, Pierre Le Bodic

Publikováno v: Journal of Artificial Intelligence Research. 72:1251-1279

Decision sets and decision lists are two of the most easily explainable machine learning models. Given the renewed emphasis on explainable machine learning decisions, both of these machine learning models are becoming increasingly attractive, as they

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::38a1591890ce3b5e2d2b468971c6f945
https://doi.org/10.1613/jair.1.12719

Tractable Explanations for d-DNNF Classifiers

Autor: Xuanxiang Huang, Yacine Izza, Alexey Ignatiev, Martin Cooper, Nicholas Asher, Joao Marques-Silva

Publikováno v: Proceedings of the Thirty-Sixth AAAI Conference on Artificial Intelligence
36th AAAI Conference on Artificial Intelligence (AAAI 2022)
36th AAAI Conference on Artificial Intelligence (AAAI 2022), AAAI: American Association for Artificial Intelligence, Feb 2022, Seattle (virtual), United States. pp.5719-5728, ⟨10.1609/aaai.v36i5.20514⟩
Proceedings of the 36th AAAI Conference on Artificial Intelligence
36th AAAI Conference on Artificial Intelligence (AAAI 2022), AAAI: Association for the Advancement of Artificial Intelligence, Feb 2022, virtual, Canada. pp.5719-5728, ⟨10.1609/aaai.v36i5.20514⟩

International audience; Compilation into propositional languages finds a growing number of practical uses, including in constraint programming, diagnosis and machine learning (ML), among others. One concrete example is the use of propositional langua

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b7de220643b2ac6040c2e83955b573f3
https://hal.science/hal-03873826/document

Propositional proof systems based on maximum satisfiability

Autor: Antonio Morgado, Maria Luisa Bonet, Joao Marques-Silva, Samuel R. Buss, Alexey Ignatiev

Publikováno v: Artificial Intelligence
Artificial Intelligence, Elsevier, 2021, 300, pp.1-59. ⟨10.1016/j.artint.2021.103552⟩
Artificial Intelligence, 2021, 300, pp.1-59. ⟨10.1016/j.artint.2021.103552⟩

International audience; The paper describes the use of dual-rail MaxSAT systems to solve Boolean satisfiability (SAT), namely to determine if a set of clauses is satisfiable. The MaxSAT problem is the problem of satisfying the maximum number of claus

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::06ae2fef8f05cf52cd43ee6681dadd67
https://hal.archives-ouvertes.fr/hal-03317630