Fabrizio Russo

Fabrizio Russo is a postdoctoral Department of Computing researcher at Imperial College London, specializing in AI systems that integrate causal reasoning and computational argumentation frameworks. His work focuses on enhancing decision-making processes through transparent causal discovery techniques and explainable machine learning architectures.

• PhD in Safe and Trusted AI (2025) from Imperial College London via UKRI CDT
• Former Head of Data Science at 4most Europe (2014-2020)
• Co-organizer of Imperial College's Explainable AI Seminars Series (2020-present)

His research combines causal inference with argumentation-based reasoning to create contestable AI systems that enable human-machine collaboration in critical decision-making scenarios. Key methodologies include:

• Constraint-based causal structure learning
• Shapley value-based feature selection
• Causal graph injection into neural networks
• Interactive argumentation frameworks

Recent publications demonstrate applications in financial risk assessment (FICO HELOC), socioeconomic modeling (Adult dataset), and housing market predictions (Boston/California datasets). His work emphasizes regulatory compliance and transparency in AI deployments.

Scientific contributions include:

• 2023 AISTATS Top-Reviewer Award
• Foundational work on causal injection techniques
• Development of argumentation-based explanation systems

As a Graduate Teaching Assistant for Introduction to Machine Learning (2021-2022), he mentored students in algorithmic foundations and supervised implementation of explainable AI systems. His GitHub repositories contain open-source implementations of causal injection frameworks.