Elias Koutsoupias
Professor of Computer Science
University of Oxford
Department of Computer Science University of Oxford Wolfson Building, Parks Road Oxford, OX1 3QD, UK Office: 368 Tel: +44 (0) 1865 610738 Email: firstname.surname@cs.ox.ac.uk |
|
Short Biography
I am a professor of computer science at the University of Oxford.
My research interests include algorithmic aspects of game theory, economics and networks, online algorithms, decision-making under uncertainty, design and analysis of algorithms, computational complexity.
I was awarded the Gödel Prize in theoretical computer science for my foundational work on the price of anarchy, a key concept in algorithmic game theory. I am also a recipient of the ERC Advanced Grant, “Algorithms, Games, Mechanisms, and the Price of Anarchy.”
Prior to my current position, I held faculty appointments at the University of California, Los Angeles (UCLA) and the University of Athens. I received my Bachelor of Science in electrical engineering from the National Technical University of Athens and my Doctor of Philosophy in computer science from the University of California, San Diego (UCSD).
Teaching
- Design and analysis of algorithms, Hilary Term 2024-25
- Design and analysis of algorithms, Hilary Term 2023-24
- Design and analysis of algorithms, Hilary Term 2022-23
- Design and analysis of algorithms, Hilary Term 2021-22
- Design and analysis of algorithms, Hilary Term 2020-21
- Probability and Computing, Hilary Term 2018-19
- Probability and Computing, Hilary Term 2017-18
- Probability and Computing, Hilary Term 2016-17
Research interests
Here are some of my papers. An almost complete list can be found at DBLP and at Google Scholar.
Price of anarchy
The concept of the price of anarchy appeared in a joint paper with Christos Papadimitriou in 1999:
- Worst-case equilibria (with Christos Papadimitriou). Its influence was recognized by the Gödel Prize in 2012.
The following papers contain some followup work on the price of anarchy of finite congestion games:
- The price of anarchy of finite congestion games (with George Christodoulou)
- On the price of anarchy and stability of correlated equilibria of linear congestion games (with George Christodoulou)
- On the performance of approximate equilibria in congestion games (with George Christodoulou and Paul Spirakis)
Coordination mechanisms is one way to try to re-design systems to improve their price of anarchy:
- Coordination mechanisms (with George Christodoulou and Akash Nanavati)
The following paper considers the interplay between price of anarchy and social inequality
- Wealth inequality and the price of anarchy (with Kurtuluş Gemici, Barnabé Monnot, Christos Papadimitriou, and Georgios Piliouras)
Mechanism design
Together with George Christodoulou and Annamaria Kovacs, we have proved the Nisan-Ronen conjecture, a major open problem in algorithmic mechanism design.
The proof and a general overview of it are in
- A proof of the Nisan-Ronen conjecture - STOC 2023 (with George Christodoulou and Annamaria Kovacs)
- A Proof of the Nisan-Ronen Conjecture - an Overview - SIGecom Exch. (with George Christodoulou and Annamaria Kovacs)
Earlier papers made progress on special cases and developed useful mechanism characterizations.
- On the Nisan-Ronen conjecture - FOCS 2021 (with George Christodoulou and Annamaria Kovacs)
- On the Nisan-Ronen conjecture for submodular valuations - STOC 2020 (with George Christodoulou and Annamaria Kovacs)
We have also established positive results for specific input instances.
- Truthful allocation in graphs and hypergraphs - ICALP 2021 (with George Christodoulou and Annamaria Kovacs)
The following are some of my earlier publications on this or related topics.
- A lower bound for scheduling mechanisms (with George Christodoulou and Angelina Vidali)
- A lower bound of 1+φ for truthful scheduling mechanisms (with Angelina Vidali)
- Mechanism design for fractional scheduling on unrelated machines (with George Christodoulou and Annamaria Kovacs)
- Scheduling without payments
Game-theoretic issues of blockchains
Blockchains raise some challenging game-theoretic questions.
The following papers study equilibria of mining games:
- Blockchain mining games (with Aggelos Kiayias, Maria Kyropoulou, and Yiannis Tselekounis)
- Energy equilibria in proof-of-work mining (with Amos Fiat, Anna Karlin, and Christos Papadimitriou)
- Fairness and efficiency in dag-based cryptocurrencies (with Georgios Birmpas, Philip Lazos, and Francisco Marmolejo-Cossio)
- Blockchain mining games with pay forward (with Philip Lazos, Paolo Serafino, and Foluso Ogunlana)
The next paper explores mechanisms for incentivising blockchain stakeholders to organize into pools
- Reward sharing schemes for stake pools (with Lars Brünjes, Aggelos Kiayias, and Aikaterini-Panagiota Stouka)
Optimal auctions
The following papers aim to design profit-maximising auctions in Bayesian settings. This is achieved through a duality framework involving partial differential inequalities, which facilitates both the construction and validation of optimal auction designs.
- Duality and optimality of auctions for uniform distributions (with Yiannis Giannakopoulos)
- Selling two goods optimally (with Yiannis Giannakopoulos)
- The FedEx problem (with Amos Fiat, Kira Goldner, and Anna Karlin)
Online algorithms
A large part of my research is about online algorithms.
The k-server conjecture is a central and highly challenging problem in online algorithms. Even after many years, this paper remains the state-of-the-art result:
- On the k-server conjecture (with Christos Papadimitriou)
The following is a review article on the k-server problem:
Here are some recent relevant papers on the k-server problem and its variants
- Towards the k-server conjecture: A unifying potential, pushing the frontier to the circle - ICALP 2021 (with Christian Coester)
- The online k-taxi problem - STOC 2019 (with Christian Coester)
- The infinite server problem - ICALP 2017 (with Christian Coester and Philip Lazos)
Another challenging online problem is the carpool fairness problem:
- Carpooling in social networks (with Amos Fiat, Anna Karlin, Claire Mathieu, and Rotem Zach)
Combining markets and online algorithms:
- Online market intermediation (with Yiannis Giannakopoulos and Philip Lazos)
Other
My research extends to several other areas. Here are a few examples.
EFX allocations on graphs
EFX envy-freeness has received considerable attention recently. The following paper considers the special case where each good can only be allocated between two specific agents.
- EFX allocations on graphs (with George Christodoulou, Amos Fiat, and Alkmini Sgouritsa)
Beyond worst-case analysis
Our chapter “Beyond Competitive Analysis,” co-authored with Anna Karlin, appears in Tim Roughgarden’s recent collection “Beyond Worst-Case Analysis” and draws partially on
- Beyond competitive analysis (with Christos Papadimitriou)
Distributed computing
Given a finite task, is there a wait-free distributed algorithm for it? With Eli Gafni, we showed that this problem is undecidable even for extremely simple cases.
- Three-processor tasks are undecidable (with Eli Gafni)
Traveling salesman problem
With Michelangelo Grigni and Christos Papadimitriou, we provided an approximation scheme for a limited case of Euclidean TSP, which was subsumed by Sanjeev Arora’s more general solution.
- An approximation scheme for planar graph TSP (with Michelangelo Grigni and Christos Papadimitriou)