Submission Deadline: 30 April 2026 View: 312 Submit to Special Issue
Prof. Dr. Dunhui Xiao
Email: xiaodunhui@tongji.edu.cn
Affiliation: School of Mathematical Sciences, Tongji University, Shanghai 200092, China
Research Interests: AI, machine learning, model reduction
Dr. Rui Fu
Email: freya_fr@tongji.edu.cn
Affiliation: School of Mathematical Sciences, Tongji University, Shanghai 200092, China
Research Interests: deep learning, computational fluid dynamics, uncertainty quantification, model reduction
Dynamical systems theory provides the foundational framework for understanding complex time-evolving phenomena across climate science, engineering, neuroscience, and ecology. Traditional approaches, however, struggle with high dimensionality, noise, and partial observability. Machine learning (ML) has emerged as a transformative tool, enabling data-driven modeling, prediction, state estimation, and control of nonlinear dynamics.
This special issue seeks advanced research at the intersection of ML and dynamical systems by promoting innovative methodologies for challenges such as long-term forecasting, chaos control, system identification, and uncertainty quantification. It fosters interdisciplinary collaboration among dynamical systems theory, applied mathematics, computer science, and applied fields including climate science, and biology, while emphasizing the tangible impact of ML-driven solutions in engineering, industrial, and scientific applications. Additionally, the issue strengthens theoretical foundations by exploring interpretability, stability, generalization, and convergence guarantees of ML methods in dynamical contexts, and identifies emerging challenges—such as data efficiency, extrapolation robustness, and scalability.
Topics of interest include, but are not limited to:
· ML applications in fluid mechanics, such as combustion, and turbulence modeling
· Machine learning for hazard detection and risk assessment
· Automatic discovery of hidden state variables in dynamical systems using ML
· Surrogate modeling and reduced-order modeling using ML
· Time series forecasting and anomaly detection in complex dynamical systems using ML
· Explainable AI and interpretability in engineering and sciences
· Koopman operator-based learning for dynamical system prediction
· Inverse problems and governing equation discovery from noisy data using ML
Login
Register
Submit a Paper
Propose a Special lssue