Zobrazeno 1 - 10
of 17
pro vyhledávání: '"Ayoobi, Hamed"'
Autor:
Leofante, Francesco, Ayoobi, Hamed, Dejl, Adam, Freedman, Gabriel, Gorur, Deniz, Jiang, Junqi, Paulino-Passos, Guilherme, Rago, Antonio, Rapberger, Anna, Russo, Fabrizio, Yin, Xiang, Zhang, Dekai, Toni, Francesca
AI has become pervasive in recent years, but state-of-the-art approaches predominantly neglect the need for AI systems to be contestable. Instead, contestability is advocated by AI guidelines (e.g. by the OECD) and regulation of automated decision-ma
Externí odkaz:
http://arxiv.org/abs/2405.10729
We propose ProtoArgNet, a novel interpretable deep neural architecture for image classification in the spirit of prototypical-part-learning as found, e.g., in ProtoPNet. While earlier approaches associate every class with multiple prototypical-parts,
Externí odkaz:
http://arxiv.org/abs/2311.15438
Feature attribution methods are widely used to explain neural models by determining the influence of individual input features on the models' outputs. We propose a novel feature attribution method, CAFE (Conflict-Aware Feature-wise Explanations), tha
Externí odkaz:
http://arxiv.org/abs/2310.20363
Neural networks (NNs) have various applications in AI, but explaining their decisions remains challenging. Existing approaches often focus on explaining how changing individual inputs affects NNs' outputs. However, an explanation that is consistent w
Externí odkaz:
http://arxiv.org/abs/2301.09559
Neural networks (NNs) have various applications in AI, but explaining their decisions remains challenging. Existing approaches often focus on explaining how changing individual inputs affects NNs' outputs. However, an explanation that is consistent w
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c65ef1d0caa92e46d3bafef44bb66b30
http://hdl.handle.net/10044/1/104585
http://hdl.handle.net/10044/1/104585
Autor:
Ayoobi, Hamed
In this thesis, we have introduced new techniques for the problems of open-ended learning, online incremental learning, and explainable learning. These methods have applications in the classification of tabular data, 3D object category recognition, a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::26c3c058b78decfe9f7d8a18c3011f07
Autor:
Ayoobi, Hamed1 (AUTHOR) hamed_ayoobi@yahoo.com, Cao, Ming2 (AUTHOR), Verbrugge, Rineke1 (AUTHOR), Verheij, Bart1 (AUTHOR)
Publikováno v:
IEEE Transactions on Automation Science & Engineering. Oct2022, Vol. 19 Issue 4, p3419-3433. 15p.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Publikováno v:
20th IEEE International Conference on Machine Learning and Applications, 1118-1123
STARTPAGE=1118;ENDPAGE=1123;TITLE=20th IEEE International Conference on Machine Learning and Applications
STARTPAGE=1118;ENDPAGE=1123;TITLE=20th IEEE International Conference on Machine Learning and Applications
Human agents can acquire knowledge and learn through argumentation. Inspired by this fact, we propose a novel argumentation-based machine learning technique that can be used for online incremental learning scenarios. Existing methods for online incre
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d937448c19433c2d01eb98acc1074bcb
https://doi.org/10.1109/icmla52953.2021.00183
https://doi.org/10.1109/icmla52953.2021.00183
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.