AI-SCI Workshop at the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD) 2025
Artificial Intelligence (AI) is transforming modern society. It has high potential to contribute to more flexible planning and operations of safety-critical infrastructures under deep transformation to tackle global challenges such as climate change, energy transition, digital transformation, and infrastructure resilience to natural and human-made hazards. However, this faces challenges, including ensuring AI trustworthiness through reliability, transparency, and robustness; addressing uncertainties from aged assets, climate change, and rising demand; enabling human-AI joint decision-making for effective collaboration; and overcoming scalability barriers of AI-based systems when applied to large-scale networks. Addressing these issues requires testing and experimentation environments and protocols for rigorous assessment to ensure safe and efficient integration into complex systems.
The main research areas discussed in this workshop include, but are not limited to, supervised and unsupervised learning, reinforcement learning, human-AI interface design, interactive AI, physics-informed machine learning, data-driven simulation, explainable AI, AI ethics and regulation, human factors in AI, and socio-technical system design.
In ECML PKDD 2025, this workshop will showcase real-world examples and provide researchers with insights into AI applications within the safety-critical infrastructure domain. Use cases cover key sectors such as electricity, transportation, water, and information and communication technologies. Additionally, the workshop will tackle designing optimal collaborations between humans and AI, focusing on how AI complements and enhances human capabilities. For industry professionals, it will highlight state-of-the-art AI-driven solutions with potential to improve efficiency and resilience in critical infrastructures.
The workshop is sponsored by the Horizon Europe AI4REALNET (“AI for REAL-world NETwork operation”) project.