Explainable Artificial Intelligence (XAI) 2.0: A manifesto of open challenges and interdisciplinary research directions

被引:77
|
作者
Longo, Luca [1 ,2 ]
Brcic, Mario [3 ]
Cabitza, Federico [4 ,5 ]
Choi, Jaesik [6 ,7 ]
Confalonieri, Roberto [8 ]
Del Ser, Javier [9 ,10 ,11 ]
Guidotti, Riccardo [12 ]
Hayashi, Yoichi [13 ]
Herrera, Francisco [11 ]
Holzinger, Andreas [14 ]
Jiang, Richard [15 ]
Khosravi, Hassan [16 ]
Lecue, Freddy [17 ]
Malgieri, Gianclaudio [18 ]
Paez, Andres [19 ,20 ]
Samek, Wojciech [21 ,22 ,23 ]
Schneider, Johannes [24 ]
Speith, Timo [25 ,26 ]
Stumpf, Simone [27 ]
机构
[1] Artificial Intelligence & Cognit Load Res Lab, Dublin, Ireland
[2] Technol Univ Dublin, Sch Comp Sci, Dublin, Ireland
[3] Univ Zagreb, Fac Elect Engn & Comp, Zagreb, Croatia
[4] Univ Milano Bicocca, Milan, Italy
[5] IRCCS Osped Galeazzi St Ambrogio, Milan, Italy
[6] Korea Adv Inst Sci & Technol KAIST, Kim Jaechul Grad Sch AI, Daejeon, South Korea
[7] INEEJI Corp, Seongnam, South Korea
[8] Univ Padua, Dept Math, Padua, Italy
[9] TECNALIA, Basque Res & Technol Alliance BRTA, Derio, Spain
[10] Univ Basque Country, UPV EHU, Bilbao, Spain
[11] Univ Granada, DaSCI Andalusian Inst Data Sci & Computat Intellig, Dept Comp Sci & Artificial Intelligence, Granada, Spain
[12] Univ Pisa, Pisa, Italy
[13] Meiji Univ, Dept Comp Sci, Tokyo, Japan
[14] Univ Nat Resources & Life Sci Vienna, Human Ctr AI Lab, Vienna, Austria
[15] Univ Lancaster, Sch Comp & Commun, Lancaster, England
[16] Univ Queensland, Brisbane, Australia
[17] INRIA, Natl Inst Res Digital Sci & Technol, Sophia Antipolis, France
[18] Leiden Univ, eLaw Ctr Law & Digital Technol, Leiden, Netherlands
[19] Univ Andes, Dept Philosophy, Bogota, Colombia
[20] Univ Los Andes, Ctr Res & Format Artificial Intelligence CinfonIA, Bogota, Colombia
[21] Tech Univ Berlin, Berlin, Germany
[22] Fraunhofer Heinrich Hertz Inst, Berlin, Germany
[23] Berlin Inst Fdn Learning & Data BIFOLD, Berlin, Germany
[24] Univ Liechtenstein, Dept Informat Syst & Comp Sci, Vaduz, Liechtenstein
[25] Univ Bayreuth, Dept Philosophy, Bayreuth, Germany
[26] Saarland Univ, Ctr Perspicuous Comp, Saarbrucken, Germany
[27] Univ Glasgow, Sch Comp Sci, Glasgow, Scotland
关键词
Explainable artificial intelligence; XAI; Interpretability; Manifesto; Open challenges; Interdisciplinarity; Ethical AI; Large language models; Trustworthy AI; Responsible AI; Generative AI; Multi-faceted explanations; -based; Causality; Actionable XAI; Falsifiability; BLACK-BOX; AI; EXPLANATIONS; CLASSIFICATION; REPRESENTATION; ARGUMENTATION; TAXONOMY; LEARNER; SYSTEMS;
D O I
10.1016/j.inffus.2024.102301
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Understanding black box models has become paramount as systems based on opaque Artificial Intelligence (AI) continue to flourish in diverse real-world applications. In response, Explainable AI (XAI) has emerged as a field of research with practical and ethical benefits across various domains. This paper highlights the advancements in XAI and its application in real-world scenarios and addresses the ongoing challenges within XAI, emphasizing the need for broader perspectives and collaborative efforts. We bring together experts from diverse fields to identify open problems, striving to synchronize research agendas and accelerate XAI in practical applications. By fostering collaborative discussion and interdisciplinary cooperation, we aim to propel XAI forward, contributing to its continued success. We aim to develop a comprehensive proposal for advancing XAI. To achieve this goal, we present a manifesto of 28 open problems categorized into nine categories. These challenges encapsulate the complexities and nuances of XAI and offer a road map for future research. For each problem, we provide promising research directions in the hope of harnessing the collective intelligence of interested stakeholders.
引用
收藏
页数:22
相关论文
共 50 条
  • [21] Explainable artificial intelligence and the social sciences: a plea for interdisciplinary research
    De Mulder, Wim
    AI & SOCIETY, 2024,
  • [22] Explainable Artificial Intelligence (XAI) for Internet of Things: A Survey
    Kok, Ibrahim
    Okay, Feyza Yildirim
    Muyanli, Ozgecan
    Ozdemir, Suat
    IEEE INTERNET OF THINGS JOURNAL, 2023, 10 (16) : 14764 - 14779
  • [23] Beauty is in the explainable artificial intelligence (XAI) of the "agnostic" beholder
    Laios, Alexandros
    De Jong, Diederick
    Kalampokis, Evangelos
    TRANSLATIONAL CANCER RESEARCH, 2023, 12 (02) : 226 - 229
  • [24] eXplainable Artificial Intelligence (XAI) in aging clock models
    Kalyakulina, Alena
    Yusipov, Igor
    Moskalev, Alexey
    Franceschi, Claudio
    Ivanchenko, Mikhail
    AGEING RESEARCH REVIEWS, 2024, 93
  • [25] eXplainable Artificial Intelligence (XAI) for improving organisational regility
    Shafiabady, Niusha
    Hadjinicolaou, Nick
    Hettikankanamage, Nadeesha
    Mohammadisavadkoohi, Ehsan
    Wu, Robert M. X.
    Vakilian, James
    PLOS ONE, 2024, 19 (04):
  • [26] Editorial: Explainable Artificial Intelligence (XAI) in Systems Neuroscience
    Lombardi, Angela
    Tavares, Joao Manuel R. S.
    Tangaro, Sabina
    FRONTIERS IN SYSTEMS NEUROSCIENCE, 2021, 15
  • [27] A Literature Review on Applications of Explainable Artificial Intelligence (XAI)
    Kalasampath, Khushi
    Spoorthi, K. N.
    Sajeev, Sreeparvathy
    Kuppa, Sahil Sarma
    Ajay, Kavya
    Maruthamuthu, Angulakshmi
    IEEE ACCESS, 2025, 13 : 41111 - 41140
  • [28] Interdisciplinary Research in Artificial Intelligence: Challenges and Opportunities
    Kusters, Remy
    Misevic, Dusan
    Berry, Hugues
    Cully, Antoine
    Le Cunff, Yann
    Dandoy, Loic
    Diaz-Rodriguez, Natalia
    Ficher, Marion
    Grizou, Jonathan
    Othmani, Alice
    Palpanas, Themis
    Komorowski, Matthieu
    Loiseau, Patrick
    Frier, Clement Moulin
    Nanini, Santino
    Quercia, Daniele
    Sebag, Michele
    Fogelman, Francoise Soulie
    Taleb, Sofiane
    Tupikina, Liubov
    Sahu, Vaibhav
    Vie, Jill-Jenn
    Wehbi, Fatima
    FRONTIERS IN BIG DATA, 2020, 3
  • [29] Explainable Artificial Intelligence for Cyber Threat Intelligence (XAI-CTI)
    Samtani, Sagar
    Chen, Hsinchun
    Kantarcioglu, Murat
    Thuraisingham, Bhavani
    IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING, 2022, 19 (04) : 2149 - 2150
  • [30] The application of explainable artificial intelligence (XAI) in electronic health record research: A scoping review
    Caterson, Jessica
    Lewin, Alexandra
    Williamson, Elizabeth
    DIGITAL HEALTH, 2024, 10