Research

• Balancing Fairness and Efficiency in Transport Network Design through Reinforcement Learning - (upcoming) - AAMAS 2023.
• Optimizing Fairness in Transport Network Design using Deep Reinforcement Learning (Dimitris Michailidis, Sennay Ghebreab and Fernando P. Santos) - (link) - IJCAI 2022, DSO Workshop.

I am a PhD student at the Civic AI Lab of the University of Amsterdam.

In my research, I use Machine Learning to design inclusive, accessible and sustainable cities.

My current focus is on using Reinforcement Learning to explore the trade-off between utility and equity/fairness when designing public transport networks.

In the near future, I would like to investigate how would urban populations adapt to such changes in a city's public tranport networks.

My goal is to create a multi-agent framework that helps to envision alternative urban designs and their evolution in time.

Sounds interesting? Then please, get in touch!

Teaching & Supervision

Master's Students Supervision

• Riccardo Fiorista (fiorista.riccardo@gmail.com) - Equitable Public Transport Network Reduction: A Reinforcement Learning Approach - MSc Thesis, Spring 2022
• Paul Stepanov (paulstepanov@yahoo.com) - Fairness in Facility Location through Clustering - MSc Thesis, Spring 2022
• Lodewijk van Keizerswaard (lodewijk.keizer@gmail.com) - Rewarding rationality to improve reinforcement learning models on the metro network design problem - MSc AI Project, Summer 2022

Courses

• Fundamentals of Data Science - Information Studies MSc - Data Science Track (2021, 2022)
• Fairness Accountability Confidentiality and Transparency in AI - Artificial Intelligence MSc MSc (2022, 2023)

Personal Projects

I am also developing Greece in Figures, an organization that aims to explain complex socioeconomic phenomena in simple words & clean graphs, for the Greek audience.

Occasionally, I create interactive data stories on society and culture and post them on my blog Thousand Words.

Communication

You can reach me by email: d.michailidis at uva.nl, on Twitter or Linkedin.