Guest Editors
Dr. Mehdi Neshat, Torrens University Australia, Australia
Dr. Menasha Thilakaratne, University of Adelaide, Australia
Dr. Yue Xie, University of Cambridge, United Kingdom
Dr. Mosab Hamdan, King Fahd University of Petroleum and Minerals, Saudi Arabia
Summary
Real-world problems present distinct challenges that distinguish them from synthetic mathematical problems: Real-world problems entail intricate systems containing multiple interdependent components, rendering their modelling and analysis more demanding. These systems frequently manifest nonlinear behaviour, dependencies, and uncertainties, necessitating advanced machine-learning techniques that can effectively capture and represent such intricacies.
Real-world applications frequently grapple with massive datasets, giving rise to computational and memory constraints. Advanced machine learning and optimization methods specifically interest in emerging and interdisciplinary methodologies, such as transfer learning, federated learning, or quantum machine learning must address these limitations in computational resources, ensuring efficiency and scalability while processing and analyzing extensive volumes of data. Meanwhile, this special issue encourages submissions that not only present advanced techniques but also demonstrate scalability and efficiency improvements over existing approaches. In addition, we extend an invitation for submissions detailing the deployment of these methods in real-world environments, including case studies that demonstrate practical benefits and challenges encountered during implementation. Considering the application of these methods in real-world scenarios, this special issue invites discussions on the societal, ethical, and regulatory implications of deploying advanced machine learning systems.
Developing techniques that can accommodate the computational demands of these large-scale datasets is of utmost importance for their practical implementation and deployment in real-world scenarios. For addressing large-scale and complex problems, the role of benchmark datasets and evaluation metrics is crucial, therefore, introducing new datasets or providing novel insights into established ones are valuable. Furthermore, real-world problems often involve conflicting objectives that must be simultaneously optimized. Multi-objective optimization techniques can aid in identifying trade-offs and generating a set of Pareto-optimal solutions, thereby providing decision-makers with a range of alternatives that strike a balance between multiple objectives. This special issue will emphasize these topics to address the main challenges encountered in complex, expensive and large-scale real problems across various industrial (e.g. energy, renewable energy systems, network optimization, vehicle routing and fleet management, production planning and scheduling), health problems (e.g., bioinformatics, drug discovery and design, disease diagnosis and biomarker discovery), and new technologies (e.g., edge computing, IoT devices) in optimizing and executing machine learning algorithms in a real-world context.
Keywords
Real-world problems, Advanced Machine learning, Deep learning, Optimization, Metaheuristic Algorithms, Evolutionary Computation, Single Objective Methods, Multi-Objective Optimization methods, Large-scale problems, Expensive optimization problems, Case studies, Operations research, Industry-specific expertise
Published Papers
Login
Register
Submit a Paper
Propose a Special lssue