Guest Editors
Assoc. Prof. Wei Zhang
Email: wei.zhang@singaporetech.edu.sg
Affiliation: Information and Communications Technology Cluster, Singapore Institute of Technology, Singapore, Singapore
Homepage:
Research Interests: artificial intelligence (AI) for industrial applications
Dr. Haiyan Yin
Email: yin_haiyan@a-star.edu.sg
Affiliation: A*STAR Centre for Frontier AI Research, Singapore, Singapore
Homepage:
Research Interests: aentic AI, reinforcement learning, meta-learning, and trustworthy decision-making
Assoc. Prof. Pan Lai
Email: plai1@ntu.edu.sg
Affiliation: College of Computer Science, South-Central Minzu University, Wuhan, China
Homepage:
Research Interests: big data and AI , resource scheduling algorithms
Summary
Reinforcement Learning (RL) has emerged as a transformative paradigm in artificial intelligence, enabling agents to learn optimal behaviors through interaction with complex environments. Its advancement is critical for developing autonomous systems that can solve real-world decision-making problems in dynamic and uncertain settings, from robotics and resource management to scientific discovery.
This Special Issue aims to showcase cutting-edge research that pushes the boundaries of traditional Reinforcement Learning. We seek original contributions that address core challenges in scalability, sample efficiency, safety, and generalization, and explore synergistic integrations with other fields. The scope encompasses novel algorithms, theoretical foundations, and significant applications that demonstrate a clear leap beyond conventional RL approaches, particularly in domains aligned with the journal's focus, such as networked systems, cybersecurity, materials science, and industrial IoT.
Suggested Themes
· Advanced RL Algorithms: Novel model-based RL, inverse RL, meta-RL, and hierarchical RL methods for improved efficiency and generalization.
· RL at the Intersection with Other Paradigms: Hybrid models integrating RL with deep learning, symbolic AI, transfer learning, or evolutionary computation.
· Safe, Robust, and Trustworthy RL: Research on risk-sensitive control, adversarial robustness, interpretability, and ethical alignment in RL systems.
· Multi-agent and Distributed RL: Algorithms for cooperation, competition, and communication in scalable multi-agent systems.
· RL for Science and Engineering Applications: Applications of RL in material discovery, network optimization, autonomous cyber-defense, and smart manufacturing.
· Theoretical Advances in RL: New frameworks for convergence analysis, sample complexity, and the exploration-exploitation trade-off.
· Efficient RL via Simulation and Digital Twins: Leveraging high-fidelity simulators and digital twins for training and deploying RL agents in physical systems.
Keywords
reinforcement learning, decision making, RL for science and engineering applications
Login
Register
Submit a Paper
Propose a Special lssue