A bit about me
I was previously a PhD Candidate at the Civic AI Lab.
My research uses multi-agent systems and machine learning to study societal problems: how the behavior of individual agents produces collective outcomes, and how those outcomes are distributed across social groups.
In my PhD I developed (multi-objective) reinforcement learning methods for designing equitable public transport networks, and agent-based models of school segregation and inequality in congestion games.
Now I study generative social simulation with LLM-based agents: how agent design choices shape collective outcomes such as polarization and network segregation, and how to make these simulations more reliable for social science and policy.
I also work on prosocial recommender systems: who-to-follow recommendations designed to reduce echo chambers rather than maximize engagement, which I study through user experiments with real Bluesky users.
Outside academia, I am the co-founder of Greece in Figures, a website that explains complex socioeconomic phenomena to a Greek audience.
If any of this sounds interesting, write to me.
Papers
-
2026
From Caricature to Collective Artifact: How Persona Construction Drives Polarization in Generative Social Simulations
Michailidis, D., Törnberg, P., Santos, F.P.
arXiv preprint arXiv:2610.16498 (2026)
-
2026
Smart Transportation Without Neurons — Fair Metro Network Expansion with Tabular Reinforcement Learning
Michailidis, D., Ghebreab, S., Santos, F.P.
arXiv preprint arXiv:2606.04167 (2026)
-
2026
Scalable Multi-Objective Reinforcement Learning with Fairness Guarantees using Lorenz Dominance
Michailidis, D., Röpke, W., Roijers, D., Ghebreab, S., Santos, F.P.
Journal of Artificial Intelligence Research 85 (2026)
- 2026
-
2026
Inequality in Congestion Games with Learning Agents
Michailidis, D., Ghebreab, S., Santos, F.P.
AAMAS 2026
-
2025
A Summary of: Tackling School Segregation with Transportation Network Interventions — An Agent-Based Modelling Approach
Michailidis, D., Tasnim, M., Ghebreab, S., Santos, F.P.
JAAMAS Track — AAMAS 2025
-
2025
PTFA: An LLM-Based Agent that Facilitates Online Consensus Building through Parallel Thinking
Gu, W., Li, Z., Buermann, J., Dilkes, J., Michailidis, D., Hasegawa, S., Yazdanpanah, V., Stein, S.
Pacific Rim International Conference on Artificial Intelligence (PRICAI 2025)
-
2025
Understanding Fairness in Congestion Games with Learning Agents
Michailidis, D., Ghebreab, S., Santos, F.P.
Adaptive and Learning Agents Workshop, AAMAS (2025)
-
2024
Where (not) to Cross the Street
van Oorschot, L., Michailidis, D., IJzerman, N., Jansen, S., Roijers, D.
BNAIC/BeNeLearn (2024)
-
2024
Enabling Inclusive Urban Transport Planning Through Civic Artificial Intelligence
Michailidis, D., Khutsishvili, K., Konstantis, K., Tympas, A., Ibrahim, I.A., Ghebreab, S.
Strengthening European Mobility Policy. Palgrave Macmillan, Cham (2024)
-
2024
Tackling School Segregation with Transportation Network Interventions: An Agent-Based Modelling Approach
Michailidis, D., Tasnim, M., Ghebreab, S., Santos, F.P.
Autonomous Agents and Multi-Agent Systems 38, 22 (2024)
Most Visionary Paper Award, CMAS-AAMAS 2023
-
2024
Towards Reducing School Segregation by Intervening on Transportation
Michailidis, D., Tasnim, M.
Autonomous Agents and Multiagent Systems. Best and Visionary Papers: AAMAS 2023 Workshops. Springer (2024)
- 2024
-
2023
Fair Transport Network Design using Multi-Agent Reinforcement Learning
Michailidis, D.
Doctoral Consortium — AAMAS 2023
- 2023
-
2023
Fairness in Transport Network Design — A Multi-Objective Reinforcement Learning Approach
Michailidis, D., Röpke, W., Ghebreab, S., Roijers, D., Santos, F.P.
Adaptive and Learning Agents Workshop, AAMAS (2023)
-
2023
Equitable Public Transit Network Reduction
Fiorista, R., Michailidis, D., Santos, F.P.
Transportation Research Board Annual Meeting (2023)
1st Place, Network Modeling Competition, TRB 2022
-
2022
Socially-Aware Artificial Intelligence for Fair Mobility
Michailidis, D.
AAAI/ACM Conference on AI, Ethics, and Society (AIES 2022)
-
2018
Real Time Location Based Sentiment Analysis on Twitter - The AirSent System
Michailidis D., Stylianou N., Vlahavas I.
10th Hellenic Conference on Artificial Intelligence (2018)
Things I wrote
Teaching and supervision
Master's Students Supervision
-
Riccardo Fiorista (LinkedIn)
Equitable Public Transport Network Reduction: A Reinforcement Learning Approach — MSc Thesis, Spring 2022 -
Paul Stepanov (LinkedIn)
Fairness in Facility Location through Clustering — MSc Thesis, Spring 2022 -
Lodewijk van Keizerswaard (LinkedIn)
Rewarding rationality to improve reinforcement learning models on the metro network design problem — MSc AI Project, Summer 2022 -
Laura Brafine (LinkedIn)
Measuring the Impact of Transportation Systems on Urban Segregation Through Random Walks — MSc Thesis, Spring 2023 -
Leah van Oorschot (LinkedIn)
Where (not) to Cross the Street: Automatically Detecting Crosswalks for Accessible Navigation — MSc Thesis, Spring 2024 -
Arno Bhattacharya (LinkedIn)
Scalable Transport Network Expansion: Applying Deep Reinforcement Learning to a Reduced State-Action Space — MSc Thesis, 2026 -
Rovshan Tamirov (LinkedIn)
Reinforcement Learning for Optimising Public Electric Vehicle Charging Station Placement in Amsterdam — MSc Thesis, 2026
Courses Taught
- Agentic Artificial Intelligence — Artificial Intelligence MSc (2026)
- Reinforcement Learning — Artificial Intelligence MSc (2023)
- Fundamentals of Data Science — Information Studies MSc, Data Science Track (2021, 2022)
- Fairness Accountability Confidentiality and Transparency in AI — Artificial Intelligence MSc (2022, 2023)
(Sub-)Reviewing
- AAMAS (2027)
- Nature Communications (2026)
- Neural Computing and Applications (2026)
- Advanced Science (2026)
- IEEE Transactions on Evolutionary Computation (2026)
- IJCAI (2026)
- Adaptive and Learning Agents Workshop (ALA), AAMAS (2026)
- Transportation Research Part C (2024)
- NetSciX (2024)
- AAAI (2024)
- ECAI (2023)
- Multi-Objective Decision Making Workshop (ECAI 2023, ECAI 2024, IJCAI-ECAI 2026)
- Public Transport Journal (2022)
Side projects
- Greece in Figures
Explaining complex socioeconomic phenomena for the Greek audience.
- 100 Europeans
Raising awareness around big challenges through interactive and visual stories.
Say hello
d.michailidis [at] uva.nl
Civic AI Lab
University of Amsterdam
Science Park, Amsterdam
Netherlands