Dr. Nicolò Pagan

Nicolò joined the Social Computing Group as a Lecturer of the “Social Computing” course in February 2021 and as Post-Doctoral Researcher in May 2021. Since January 2022, he is also a member of the National Centre of Competence in Research «Dependable, ubiquitous automation» (NCCR Automation).

In April 2021, he received his Ph.D. from ETH Zürich advised by Prof. Florian Dörfler, defending his dissertation titled “Modelling, Analysis, and Inference in Social Network Formation”.

Nicolò also holds a M.Sc. in Computational Science and Engineering from EPF Lausanne, a B. Sc. in Applied Mathematics from Politecnico di Torino (Italy), and he worked in the Computational Fluid-dynamics domain as a software engineer and research consultant at Ascomp AG, a spin-off of ETH Zürich.

Nicolò's primary research interests lie in the realm of AI ethics, where he explores the complex interplay between technology and human behavior in an intricate network system. Through his expertise in mathematical modeling, complex systems analysis, and data science, he is able to delve into this field of study with a unique perspective and he aims to contribute valuable insights to the ongoing discourse on the ethical implications of AI and technology in our rapidly evolving digital landscape.

More precisely, his current research revolves around two pillars:

• the long-term fairness effect of feedback loops engendered by automated Machine Learning-based decision-making systems.
• the use of generative AI, based on Large Language Models (LLMs), for building large-scale human environments for testing health notification interventions.

In his previous work, he investigated the impact of recommender systems within social media platforms, with a particular emphasis on discerning how these systems influence the fairness of the network formation processes. Two of his recent works have been published in Nature Communications.

More broadly speaking, Nicolò is interested in Social and Engineering Systems and in problems that correspond to significant societal and ethical challenges, with an emphasis on areas such as autonomous systems, energy systems, finance, social networks, urban systems, and their underlying interconnections.

2023

• A. Bouleimen, N. Pagan, S. Cresci, A. Urman, and S. Giordano: Dynamics of toxic behavior in the Covid-19 vaccination debate. Forthcoming at the 12^th International Conference on Complex Networks and their Applications (2023).
• N. Lanzetti, F. Dörfler, and N. Pagan: The Impact of Recommendation Systems on Opinion Dynamics: Microscopic versus Macroscopic Effects, Forthcoming at the 62^nd IEEE Conference on Decision and Control (CDC) (2023).
• N. Pagan*, J. Baumann*, E. Elokda, G. De Pasquale, S. Bolognani, and A. Hannák: A Classification of Feedback Loops and Their Relation to Biases in Automated Decision-Making Systems. Forthcoming at the 3^rd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO ’23).
• S. Ionescu, A. Hannák, and N. Pagan: Group Fairness for Content Creators: the Role of Human and Algorithmic Biases under Popularity-based Recommendations 17^th ACM Conference on Recommender Systems 2023.
• S. Ioneuscu, A. Hannák, and N. Pagan: The Role of Luck in the Success of Social Media Influencers, Applied Network Science (2023).
• A. Maddux, N. Pagan, G. Belgioioso, and F. Dörfler: Data-Driven Behaviour Estimation in Parametric Games, 22^nd World Congress of the International Federation of Automatic Control (2023).

2022

• S. Ionescu S, N. Pagan, A. Hannak: Individual Fairness for Social Media Influencers. 11^th International Conference on Complex Networks and their Applications 2022.

2021

• N. Pagan, W. Mei W, C. Li, F. Dörfler:A meritocratic network formation model for the rise of social media influencers. Nature Communications. 2021 Nov 30;12(1):1-2.

2019

• N. Pagan, F. Dörfler: Game theoretical inference of human behavior in social networks. Nature communications. 2019 Dec 3;10(1):1-2.

• 2021: Lecturer of the Social Computing course.

• Fairness-efficiency trade-off in ride-hailing services
• Closed-loop between Opinion Dynamics and Recommender Systems
• Re-Tweetted or Re-tweeeter: who benefits the most?

See the full list here.

• 10/2023 - present: Jiaqi Zhang, University of Zurich. Master Thesis: Using generative AI for modelling large-scale human behaviour for health notification interventions
• 03/203 - 08/2023 Michael Blum, University of Zurich. Independent Study: How Specialization Shapes Pattern Recognition: Empirical Evidence from the Domain of Chess.
• 05/2022 - 05/2023. Maria Helena Margareta Pelli and Andrianos Michail, University of Zürich. Master Project: “A longitudinal study of emotions on Twitter chains of re-tweets”.
• 03/2022 - 07/2022.  Jules Authier, ETH Zürich. Semester Project: “Analysis of the Closed-Loop between Recommender Systems and Opinion Dynamics”

• 06-2021 - 11/2021. Michael Blum, University of Zürich. Bachelor Thesis: “Learning and success in chess: the role of openings patterns”.