Home
I am a tenure-track associate professor in the Machine Learning Group at the University of Bergen and a honorary associate professor at the University of Warwick. My current research interests focus on the study of structural causal models and causal abstraction, as well as on methodologies for learning and exploiting abstraction in machine learning and reinforcement learning.
Previously, I worked as as a post-doctoral research asssociate in the Warwick Machine Learning Group at the University of Warwick, where I led research on machine learning, causality and abstraction in the group of Prof Theo Damoulas. During my first postdoc, I did research on reinforcement learning, computer security, and uncertainty in the Information and Cyber Security group at the University of Oslo under Prof Audun Josang. As a PhD, I studied unsupervised learning and information theoretic learning under the supervision of Prof Ke Chen at the University of Manchester.
I am also interested in the systematization of machine learning, and in the potential intersections of machine learning with other topics such as physics, politics, economics and philosophy.
Highlighted papers
- Joel Dyer, Nicholas Bishop, Yorgos Felekis, Fabio Massimo Zennaro, Anisoara Calinescu, Theodoros Damoulas, Michael Wooldridge. Interventionally Consistent Surrogates for Agent-based Simulators, Published in NeurIPS 2024 [Blogpost]
- Fabio Massimo Zennaro, Nicholas Bishop, Joel Dyer, Yorgos Felekis, Anisoara Calinescu, Michael Wooldridge, Theodoros Damoulas. Causally Abstracted Multi-armed Bandits, Published in UAI 2024 [Oral presentation]
- Yorgos Felekis, Fabio Massimo Zennaro, Nicola Branchini, Theodoros Damoulas. Causal Optimal Transport of Abstractions, Published in CLeaR 2024
- Fabio Massimo Zennaro. Abstraction between Structural Causal Models: A Review of Definitions and Properties, Published in UAI 2022 Workshop on Causal Representation Learning [Best paper award]
- Fabio Massimo Zennaro, Paolo Turrini, Theodoros Damoulas. Quantifying Consistency and Information Loss for Causal Abstraction Learning, Published in IJCAI 2023
- Fabio Massimo Zennaro, Máté Drávucz, Geanina Apachitei, W. Dhammika Widanage, Theodoros Damoulas. Jointly Learning Consistent Causal Abstractions Over Multiple Interventional Distributions, Published in CLeaR 2023 [Oral presentation: 9% acceptance rate]
Highlighted talks
- Abstraction between Structural Causal Models and Measures of Abstraction Error @ University of Kyoto 2023
- A Gentle Introduction to Casual Models @ OsloMet AI Seminar 2019
- Applications of reinforcement learning to computer security: problems, models, and perspectives @ OsloMet AI Seminar 2021
- A Left Realist Critique of the Political Value of Adopting Machine Learning Systems in Criminal Justice @ ECML Workshop on AI for Social Good (SoGood) 2020
- Information Bottleneck (and Unsupervised Learning), @ Robotics and Intelligent Systems (ROBIN) Seminar 2020