Machine Learning

Machine Learning (ML) has evolved from a subfield of Artificial Intelligence (AI) into a major research discipline underpinning contemporary information processing, data analytics, knowledge discovery, and intelligent decision-making. The rapid growth of large-scale, heterogeneous, multimodal, streaming, and distributed data has created new opportunities to develop learning-based methods that extract knowledge and support complex decisions across scientific, industrial, and societal domains.

Recent advances in deep learning, self-supervised and continual learning, foundation models, multimodal and graph learning, federated learning, generative models, and causal learning are significantly expanding the capabilities of ML. At the same time, real-world deployment has highlighted critical challenges concerning explainability, robustness, uncertainty, fairness, privacy, security, reproducibility, efficiency, sustainability, and human–AI collaboration. These developments motivate the establishment of a dedicated Machine Learning Section in Informatics to provide a focused scholarly platform connecting methodological advances with information-centric applications.

The proposed Section will cover the complete machine learning lifecycle, including data preparation and representation, learning algorithms and architectures, model training and adaptation, evaluation and benchmarking, deployment, monitoring, and continuous learning. It will welcome both foundational methodological contributions and rigorous application-oriented research, particularly work addressing complex, dynamic, heterogeneous, and human-centred environments.

Scope

The Section will cover, but is not limited to, the following:

• Supervised, unsupervised, semi-supervised, and self-supervised learning;
• Statistical, probabilistic, deep, representation, and reinforcement learning;
• Transfer, few-shot, continual, lifelong, and meta-learning;
• Federated, distributed, decentralized, online, and streaming ML;
• Graph machine learning and graph neural networks;
• Foundation models, generative ML, large language models, and multimodal learning;
• Neuro-symbolic, causal, retrieval-augmented, and knowledge-enhanced learning;
• Data-centric ML, synthetic data, anomaly detection, feature learning, and learning from noisy, incomplete, or evolving data;
• Explainable, interpretable, robust, fair, privacy-preserving, secure, and responsible ML;
• Uncertainty quantification, trustworthy prediction, and human-centred ML;
• ML systems, MLOps, edge/IoT ML, scalable deployment, monitoring, and sustainable ML;
• Applications in manufacturing, Industry 4.0, transportation, healthcare, energy, cybersecurity, finance, education, environmental systems, scientific and engineering domains, and intelligent information systems.