Control, Learning, and Security for Scalable Information Systems
SCOPE AND AIMS
In the era of the Internet of Everything, information systems are undergoing a paradigm shift toward unprecedented scale, heterogeneity, and complexity. From cloud-edge continuums and industrial IoT to smart cities and intelligent transportation networks, modern information systems are no longer static or isolated; they are dynamic, distributed, and deeply interconnected. While this scalability offers immense potential for operational efficiency and data utility, it simultaneously presents formidable challenges that traditional methodologies cannot address in isolation. The primary rationale behind this Special Issue is to explore the critical intersection of Control, Learning, and Security, arguing that the resilience and intelligence of future scalable systems depend on the synergistic integration of these three pillars. First, the sheer scale of modern systems renders centralized control strategies impractical due to communication latency and computational bottlenecks. Consequently, there is an urgent need for distributed, robust, and event-triggered control architectures capable of managing massive numbers of agents under resource constraints. However, model-based control often struggles with the high-dimensional, unstructured data inherent in these systems. This necessitates the integration of learning methods. Artificial intelligence provides the adaptability required to handle uncertainty and optimize performance in real-time. Yet, the introduction of learning algorithms into physical systems raises new concerns: "black-box" models often lack the stability guarantees and interpretability required for safety-critical applications. Therefore, developing learning-based control strategies that are theoretically sound and computationally efficient is a key focus of this issue. Furthermore, as systems become more open and intelligent, they inevitably expose larger attack surfaces. Security is no longer an add-on but a fundamental design constraint. Scalable information systems face sophisticated threats, ranging from adversarial attacks on machine learning models to false data injection attacks on control loops. The convergence of these fields introduces complex trade-offs; for instance, how to maintain privacy without degrading control performance, or how to design control protocols that remain resilient even when subsystems are compromised.
TOPICS:
1. Control strategies for scalable information systems, including feedback control, optimal control, robust control, and distributed control, as well as fault diagnosis and fault-tolerant control.
2. Learning methods for scalable information systems, including reinforcement learning, deep learning, and online learning, as well as transfer learning, multi-agent learning, and learning in uncertain environments.
3. Security protocols in scalable information systems, including the development of secure and privacy-preserving protocols with learning techniques for scalable information systems, as well as the evaluation of their effectiveness.
4. Human-machine interaction in scalable information systems, including the design of intuitive interfaces for control and monitoring of scalable information systems, as well as the evaluation of their usability and effectiveness.
5. Real-time optimization and control for scalable information systems, including the development of real-time optimization algorithms and their integration with control strategies for scalable information systems.
6. Distributed learning and edge intelligence for scalable information systems, including federated learning, decentralized optimization, and split learning across edge devices, as well as the management of communication overhead and latency.
7. Game-theoretic approaches for control and security, including non-cooperative games, stackelberg games, and mean-field games for scalable information systems, as well as incentive mechanism design for multi-agent cooperation.
8. Adversarial machine learning in scalable information environments, including the analysis of evasion and poisoning attacks on learning models, the development of robust training algorithms, as well as the certification of robustness against adversarial perturbations.
9. Sensing and actuation in scalable information systems, including the design and optimization of sensors and actuators for scalable information systems, as well as their integration with control strategies and learning techniques.
10. Formal verification and validation of learning-enabled systems, including safety verification, reachability analysis, and contract-based design for control and learning algorithms, as well as testing methodologies for scalable software architectures.
IMPORTANT DATES
l Manuscript submission deadline: Aug-10-2026
l Notification of acceptance: Sep-10-2026
l Submission of final revised paper: Oct-10-2026
l Publication of special issue (tentative): Dec-10-2026
Main Guest Editor: Qingguo Lü, Chongqing University, China, qglv@cqu.edu.cn
Guest Editor: Huaqing Li, Southwest University, China, huaqingli@swu.edu.cn
