Game Theory and Social Choice

Strategic Cost Selection in Participatory Budgeting (NeurIPS-2025)

With Piotr Faliszewski, Łukasz Janeczko, Andrzej Kaczmarczyk, Piotr Skowron, Stanisław Szufa, and Mateusz Szwagierczak

In this work we explore the ways in which project proposers can act strategically to see their submission selected with the highest possible cost (to ensure high-quality development of the project). We focus on how different voting rules shape the structure of Nash equilibria in games based on this type of strategizing.

---

Neighborhood Stability in Assignments on Graphs (WINE-2025)

With Haris Aziz, Mashbat Suzuki, and Jeremy Vollen

In this paper we explore swap stable assignments on graphs, where agents can only exchange their positions with their direct neighbors. We show the surprising existence of stable solutions on classes of graphs such as paths or cycles.

---

The Cost Perspective of Liquid Democracy: Feasibility and Control (AAAI-2025)

With Shiri Alouf-Heffetz, Łukasz Janeczko, and Giorgios Papasotiropoulos

In this paper we analyze the methods of optimal selection of voters casting a ballot in the context of liquid democracy systems. In particular, we take into account how demanding it is for the individuals to vote.

---

Swap-Stablility in Refugee Housing: A Story About Anonymous Preferences (EUMAS-2025)

With Simon Schierreich

Here, we explore the properties of methods of allocating refugees to houses in a partially filled community. We focus on the property of swap stability, showing computational complexity aspects of finding stable solutions.

---

Strategic Nominee Selection in Tournament Solutions (EUMAS-2023)

Here, I consider computational complexity aspects of games between teams that can field their representatives to compete in a tournament. I show the hardness of finding equilibrium states from the perspective of several methods of winner determination.

---

Equilibrium Computation for Knockout Tournaments Played by Groups (AAMAS-2022)

With M.S. Ramanujan and Paolo Turrini

In this paper we explore the computational aspects of strategic nominee selection by teams participating in knockout tournaments. Interestingly, we observe that the knockout-tournaments format allows for easier Nash equilibrium computation.

---

A Hotelling-Downs Framework for Party Nominees (AAMAS-2021)

With Paul Harrenstein, M.S. Ramanujan, and Paolo Turrini

Here, we introduce a game-theoretic framework aimed at capturing strategic behavior of parties in primary elections. To do that, we provide an adaptation of the classical Hotelling-Downs model and show algorithmic aspects of equilibrium computation in this setting.

---

Voting Theory

---

Guide to Numerical Experiments on Elections in Computational Social Choice (IJCAI-2024)

With Niclas Boehmer, Piotr Faliszewski, Łukasz Janeczko, Andrzej Kaczmarczyk, Simon Rey, Dariusz Stoliczki, Stanisław Szufa, and Tomasz Wąs

In this survey we provide an overview of experimental research on social choice mechanisms. We collect statistics about types of results present in the literature and suggest methodological recommendations.

---

Discovering Consistent Subelections (AAMAS-2024)

With Jerome Lang, Łukasz Janeczko, and Stanisław Szufa

Here, we study the possibility of finding coherent groups of voters based solely on the preferences over candidates that they submit. In particular, we look for groups of like-minded voters of highly conflicted communities.

---

Social Network Analysis

---

A Complexity-Theoretic Analysis of Majority Illusion in Social Networks (Journal of Artificial Intelligence Research, also AAAI-2023)

With Umberto Grandi, Lawqueen Kanesh, M.S. Ramanujan, and Paolo Turrini

We introduce research on theoretical aspects of majority illusion, i.e., the phenomenon where individuals perceive the minority opinion as the majority view. We explore both the structural properties of networks that prohibit the majority of illusion from happening, as well as methods of reducing the extent to which illusion is present in a network.

---

Convergence of Opinion Diffusion is PSPACE-Complete (AAAI-2020)

With Dmitry Chistikov, Mike Paterson, and Paolo Turrini

In this paper we explore the computational complexity of checking the convergence of synchronous opinion diffusion protocols in directed networks. We show that, in contrast to undirected networks, the problem is not tractable.

---

Abstract Argumention

---

Aggregation in Value-Based Argumentation Frameworks (TARK-2018)

With Sylvie Doutre and Umberto Grandi

In this paper, we employ voting mechanisms to abstract argumentation frameworks with argument strength based on the relative strength of values they appeal to (which might differ between audiences).