ANTIDOTE
![](http://ixa.si.ehu.es/sites/default/files/styles/large/public/anitdote.png?itok=B5MAZHhM)
![](http://ixa.si.ehu.es/sites/default/files/styles/medium/public/antidote-logo.png?itok=ucdi9TAV)
Deskribapen motza, derrigorrezkoa proiektuak logorik ez badu (eu):
Antidote (PCI2020-120717-2) MCIN/AEI /10.13039/501100011033 eta Europar Batasunaren NextGenerationEU/PRTR -ek finantzatuta
ANTIDOTEk adimen artifizial azalgarriaren ikuspegi integratua sustatzen du, non ikasketa sakonaren behe-mailako ezaugarriak giza argudiatze-gaitasunaren maila handiagoko eskemekin konbinatzen baitira.
Deskribapen motza, derrigorrezkoa proiektuak logorik ez badu (en):
Antidote (PCI2020-120717-2) funded by MCIN/AEI /10.13039/501100011033 and by European Union NextGenerationEU/PRTR
Deskribapena (en):
ANTIDOTE fosters an integrated vision of explainable AI, where low level characteristics of the deep learning process are combined with higher level schemas proper of the human argumentation capacity.
Providing high quality explanations for AI predictions based on machine learning requires combining several interrelated aspects, including, among the others: selecting a proper level of generality/specificity of the explanation, considering assumptions about the familiarity of the explanation beneficiary with the AI task under consideration, referring to specific elements that have contributed to the decision, making use of additional knowledge (e.g. metadata) which might not be part of the prediction process, selecting appropriate examples, providing evidences supporting negative hypothesis, and the capacity to formulate the explanation in a clearly interpretable, and possibly convincing, way.
Accordingly, ANTIDOTE will exploit cross-disciplinary competences in three areas, i.e. deep learning, argumentation and interactivity, to support a broader and innovative view of explainable AI. Although we envision a general integrated approach to explainable AI, we will focus on a number of deep learning tasks in the medical domain, where the need for high quality explanations, both to clinicians and to patients, is perhaps more critical than in other domains.
Deskribapen motza, derrigorrezkoa proiektuak logorik ez badu (es):
Antidote (PCI2020-120717-2) financiado por MCIN/AEI /10.13039/501100011033 y por la Unión Europea NextGenerationEU/PRTR
Kode ofiziala:
PCI2020-120717-2
Ikertzaile nagusia:
Rodrigo Agerri
Erakundea:
CHIST-ERA - INT-Acciones de Programación Conjunta Internacional (MINECO) 2020
Saila:
LSI
Hasiera data:
2021/04/01
Bukaera data:
2024/03/31
Taldeko ikertzaile nagusia:
Rodrigo Agerri
Ixakideak:
Besteak:
Agustin Martinez-Ibargüen, Elisa Espina
Kontratua:
No
Webgunea: