Guest Editors
Dr. Fa Zhu, Nanjing Forestry University, China
Dr. Xiaochun Cheng, Swansea University, UK
Dr. Nick Papanikolaou, Democritus University of Thrace, Greece
Dr. Celestine Iwendi, University of Bolton, UK
Dr. Chinmay Chakraborty, Birla Institute of Technology, Jharkhand, India
Summary
Machine learning (ML) technology has significantly promoted the development of complex systems. However, traditional learning models are obtained under the assumption that the data are independent and identically distributed (i.i.d.), which may fail in complex systems. For instance, in the Internet of Medical, the patient data may be from different physiological signals (electroencephalogram, electromyography, or photoplethysmography); in the Internet of Vehicles (IoV) or Intelligent Transportation, the images or videos may be collected under different conditions (lighting, weather, angle, etc.). In complex systems, the data from different agents may follow heterogonous distributions. How to learn ML models from a source domain to fit a target domain has become a challenging topic in complex systems. In general, the source domain and target domain have different distributions. On the other hand, with the development of complex systems, there are massive historical data. It is another issue to reuse the ML model from historical data or make the ML model adapt to current data. This special issue will focus on the advances of transfer learning in the community of machine learning to facilitate complex systems.
This special issue will provide a chance to connect researchers and practitioners to share their state-of-the-art discoveries about transfer learning for complex systems, such as transfer learning in the Internet of Things (IoT), transfer learning in traffic and environment systems, transfer learning in social networks, transfer learning in financial systems, and transfer learning in information systems.
Keywords
This special issue seeks researchers from both academia and industry to contribute their cutting-edge works and achievements. The original papers are followed on topics of interest that include, but are not limited to, the following:
• Novel algorithms and schemes of transfer learning for complex systems
• Advances of domain adaptation for complex systems
• Advances of transfer learning in information systems
• Advances of transfer learning in financial systems
• Advances of transfer learning in environment systems
• Advances of transfer learning in social networks
• Advances of transfer learning in Internet of Things
• Pre-trained model migration for intelligent applications in complex systems
• AI-inspired algorithms and mechanisms for intelligent applications in complex systems
• Security and privacy protection through transfer learning in complex systems
Login
Register
Submit a Paper
Propose a Special lssue