Explainable Computing, Modeling & Data Science in Complex Systems

Background

The rapid advancement of computational resources, spanning both hardware and software, has transformed research across many complex systems domains. Computational approaches now represent a fundamental pillar of modern scientific inquiry, enabling more efficient investigation of complex phenomena and fostering collaboration between experimental, theoretical, and computational researchers.

Advances in modelling, simulation, and artificial intelligence have substantially increased the realism, scale, and predictive capabilities of computational methods. Complex systems, ranging from biological and ecological systems to social and technological networks, can increasingly be studied through multi-scale computational frameworks informed by large and heterogeneous datasets.

At the same time, there is growing recognition that computational methods should not only provide predictive power, but also support interpretability and mechanistic understanding. Explainable computational approaches, including mechanistic modelling, interpretable AI, and hybrid data-driven methods, are therefore becoming increasingly important.

This symposium aims to bring together researchers working on computational and AI-assisted approaches to complex systems in order to present ongoing work, exchange ideas, discuss methodological challenges, and foster interdisciplinary collaboration.

Goal/Rationale

The goal of this research topic is to present, discuss, and exchange ideas on computational approaches for the study of complex systems that go beyond purely predictive performance. Given the increasing importance of explainability, interpretability, and mechanistic understanding, we are particularly interested in approaches that address current limitations of black-box AI techniques and support more transparent and trustworthy scientific modelling.

The meeting aims to encourage open discussion of current methodological challenges, existing gaps, and opportunities for interdisciplinary collaboration across areas such as biological, ecological, social, technological, and economic complex systems.

A particular focus will be on platforms, frameworks, and software tools that facilitate explainable model generation, simulation, comparison, and validation. Ideally, these approaches should support reproducibility, extensibility, and collaborative development through open-source practices and shared computational infrastructures.

Ultimately, we envision this meeting as a stepping stone toward broader international collaboration, community building, and future joint funding initiatives in explainable and computational complex systems research. More information will be made available in the near future on https://www.combynelab.com/home/news/2ndcomosys.

Scope

This symposium welcomes contributions from researchers, practitioners, and postgraduate students working on computational, modelling, and data-driven approaches to complex systems. We particularly encourage submissions that address explainability, interpretability, and mechanistic understanding in computational and AI-assisted research workflows. Topics of interest include, but are not limited to:

• Explainable artificial intelligence and interpretable machine learning
• Computational modelling and simulation of complex systems
• Hybrid mechanistic and data-driven modelling approaches
• Multi-scale and network-based modelling frameworks
• Data science methods for biological, ecological, social, and technological systems
• Model validation, uncertainty quantification, and reproducibility
• Open-source platforms, collaborative computational infrastructures, and scientific software tools
• AI-assisted scientific discovery and decision-support systems
• Ethical, transparent, and trustworthy computational methods

The symposium aims to foster interdisciplinary exchange, methodological discussion, and future collaboration across the broad field of explainable computational complex systems research.