![]() |
| |||||||||||||||
VSI: AMLDFIS 2023 : Special Issue of Information Fusion (Elsevier): New Trends of Adversarial Machine Learning for Data Fusion and Intelligent System | |||||||||||||||
Link: https://www.sciencedirect.com/journal/information-fusion/about/call-for-papers | |||||||||||||||
| |||||||||||||||
Call For Papers | |||||||||||||||
Machine learning has witnessed remarkable success in applications across a broad spectrum of academia and industries. Notably, the adversarial characteristic naturally lies in many machine learning application techniques. Adversarial Machine Learning (AML), a bleeding-edge technique that attempts to fool machine learning method and systems by generating deceptive data with imperceptible perturbations, is becoming a growing threat in artificial intelligence and machine learning fields.
AML is concerned with the deliberate design of machine learning models that can identify security challenges, perceive attacker capabilities, estimate the level of uncertainty, and understand the consequences of malicious adversaries. Particularly, the existence of adversarial examples destroys the reliability and robustness of deep neural networks, which almost impede the practical deployment of deep learning models. It has been well demonstrated that the existing deep learning models are vulnerable to carefully crafted attacks from malicious adversaries. Moreover, such a security challenge goes well beyond the simple vision systems. Robotics and control systems, communication, networking and broadcast systems, cybersecurity, and medical diagnosis systems could all potentially be subject to adversarial attacks. In presence of this, recent years have seen an emerging surge of literature on AML. AML is an emerging trend in for the data fusion and intelligent system. With the development of such technologies, the variety and quantity of data and system put higher requirements on how to use better the data and the models on adversarial machine learning frameworks and applications. This special issue mainly focuses on the recent advancements in AML models and methods, especially for attack and defense techniques in data fusion and intelligent systems; methodologies and algorithms to solve this main issue is nowadays more and more required. In this perspective, there are many challenges to be addressed in processing heterogeneous data from multiple data fusion and intelligent system on adversarial machine learning, especially for attack and defense techniques such as universality, and data fusion results’ stability. Therefore, it is a very high demanding research task to develop preferable advanced attack and defense mechanisms for exploring the weaknesses of modern machine learning and deep learning architectures in data fusion and intelligent systems. This theme issue aims to explore the state-of-the-art methodologies and applications related to all aspects of adversarial machine learning for data fusion and intelligent systems. We would like to bring together researchers and practitioners in both academic and industrial communities to approach the new trends of adversarial machine learning architectures in the field of emerging data fusion and intelligent systems, such as data fusion and reasoning, attack and defense for intelligent fusion systems, cybersecurity systems, deep learning medical diagnosis, automatic driving systems, and human-centered intelligent robots, etc. Topics of interest include, but are not limited to: 1. Coupled adversarial deep learning model for fusion classification 2. Unsupervised machine learning adversarial network for image fusion 3. Multimodal deep learning adversarial model for scalable data fusion 4. Generative adversarial learning network for multi-focus image fusion 5. Adversarial machine learning method by the fusion of class-inherent transformations 6. Multi-view representation adversarial learning for data fusion and reasoning 7. Ensemble adversarial learning for knowledge fusion 8. Generative adversarial network for medical image fusion applications 9. Attention-positive generation deep learning model for intelligent fusion systems 10.Adversarial attacks and defenses for vision fusions systems (classification, clustering, identification, detection, retrieval) 11.Adversarial attacks and defenses in multimedia fusion systems (information retrieval,cross-media systems, and multimodal systems) 12.Adversarial attacks and defenses for autonomous vehicle intelligent system 13.Adversarial machine learning techniques on multi-granularity fusion learning 14.Adversarial federated learning model in graph fusion learning system 15.Adversarial attacks and defenses for human-centered intelligent robots 16.Adversarial attacks and defenses in medical data fusion and intelligent analysis 17.Attack and defense method based on attention mechanism with adversarial samples 18.Visible image fusion synthesis with adversarial deep machine networks 19.Reliability and robustness of deep neural networks with adversarial examples 20.Adversarial attacks and defenses fusion methods for generating adversarial examples 21.Optimization algorithms in adversarial attacks and defenses 22.Security issues in adversarial deep learning-based database systems for data fusion and intelligent system 23.Real-world applications of adversarial machine learning in different data fusion and intelligent system We highly recommend the submission of supplementary multimedia files associated with each article, as it significantly increases the visibility, downloads, and citations of articles. Guest Editors Weiping Ding, School of Information Sciences and Technology, Nantong University, China Email: dwp9988@163.com Zheng Zhang, School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, China Email: zhengzhang@hit.edu.cn Luis Martínez, Computer Sciences Department, Universidad de Jaén, Spain Email: martin@ujaen.es Yu Huang, National Engineering Research Center for Software Engineering, Peking University, China Email: hy@pku.edu.cn Zehong (Jimmy) Cao, STEM (Infomation Technology), University of South Australia, Australia Email: Jimmy.Cao@unisa.edu.au Jun Liu, Ulster University, United Kingdom Email: j.liu@ulster.ac.uk Abhirup Banerjee, Department of Engineering Science, University of Oxford, Oxford, United Kingdom Email: abhirup.banerjee@eng.ox.ac.uk |
|