Special Session Organized by

Christoph-Alexander Holst, inIT – Institute Industrial IT, Germany, Volker Lohweg, inIT – Institute Industrial IT, Germany, Rui Pinto, Faculdade de Engenharia da Universidade do Porto, Portugal Simona Salicone, Politecnico di Milano, Italy, and Pedro Torres,Instituto Politécnico de Castelo Branco, Portugal.

---

Download Call for Papers

Click here to download the special session cfp.

---

Focus

Machine learning often relies on large datasets, but real-world applications frequently face data scarcity due to limited sensor capabilities, expensive labeling, or rare phenomena such as machine faults. The goal in dealing with scarce data must be to obtain as much information as possible from the available data. Overcoming data scarcity is not only a technical challenge but also a step towards sustainable AI, enabling the development of effective models with fewer resources and minimising energy-intensive data collection and labeling processes. Methods to address data scarcity include data efficient algorithms, data augmentation, transfer learning, and information fusion that explicitly model and manage uncertainty.

---

Topics under this session include:

Data-efficient Machine Learning Algorithms

Applications of Machine Learning with Scarce Data

Machine Learning for Tabular Data

Sustainable Artificial Intelligence in the Context of Scarce Data

Augmented and Transfer Learning

Synthetic Data Generation for Scarce Data Scenarios

Information Fusion in the Context of Scarce Data

Evaluation Metrics for Scarce Data Applications

Multi-sensory Systems for Data Acquisition

Data Spaces and Lakes in Smart Manufacturing

Data Scarcity Challenges in Educational AI and Learning Analytics

Information Fusion for Adaptive Learning and AI-Tutoring Systems

Uncertainty Modelling

Optimisation under Epistemic Uncertainty

Informed Machine Learning for Uncertainty

Cyber-Physical Systems for Scarce Data