Computational approaches to Explainable Artificial Intelligence: Advances in theory, applications and trends
Вычислительные подходы к объяснимому искусственному интеллекту: достижения в теории, приложениях и тенденциях
2023-07-29
SCID: 54.1/yzn76azb
Discuss with AI
Deep LearningExplainable Artificial Intelligencecomputer visionneuroimagingrobotics
Figures from the paper
Abstract (AI)
Groundbreaking advances in theoretical and applied Artificial Intelligence (AI). DL algorithms are groundedin non-linear and complex artificial neural systems that progressively extract higher-level features from data.DL is frequently compared with human-level performance in real-world tasks, such as clinical diagnostics. Itis also a turning point in solving otherwise intractable problems in the design of virtual agents or robotics; orin providing novel insight into disease progress in genomics and Neuroimaging (NI), understanding humanbrain function, the automation of industrial processes based on computer vision, and so on. In this paper, themost relevant advances from the last few years in AI and several applications to neuroscience, neuroimaging,computer vision, and robotics are presented, reviewed and discussed. In this way, we summarize the state-ofthe-art in AI methods, models and applications within a collection of works presented at the 9th InternationalConference on the Interplay between Natural and Artificial Computation (IWINAC). The works presentedin this paper are excellent examples of new scientific discoveries made in laboratories that are successfullytransferred to real-life applications.
Key Findings
1
DL has achieved human-level performance in real-world tasks, including clinical diagnostics.
2
Deep Learning (DL) excels at extracting high-level features from complex, non-linear data using artificial neural systems.
3
Recent advances in AI have enabled solutions to previously intractable problems across robotics, genomics, neuroimaging, computer vision, and industrial automation.
4
The paper reviews and summarizes state-of-the-art AI methods, models, and applications presented at the 9th IWINAC conference.
5
Works presented demonstrate successful transitions of new scientific discoveries from laboratories to real-life applications.
Research Object
Computational approaches to Explainable Artificial Intelligence (XAI) and related Deep Learning/AI methods
Research Subject
Advances in theory, applications, and trends of XAI/DL methods, including their application to neuroscience, neuroimaging, computer vision, and robotics
Publication Details
Publication Date
2023-07-29
Journal
Publisher
ISSN
Cited by
252
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai10
Gradient-based learning applied to document recognition1998
Generative adversarial networks2020
A survey on Image Data Augmentation for Deep Learning2019
Proceedings of the 24th international conference on Machine learning2007
Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining2016
Continual lifelong learning with neural networks: A review2019
Mild Cognitive Impairment2009
AI applications to medical images: From machine learning to deep learning2021
A Survey of Methods for Explaining Black Box Models2018
Computer- and Robot-Assisted Therapies to Aid Social and Intellectual Functioning of Children with Autism Spectrum Disorder2019