Georgios Piliouras received his PhD in Computer Science from Cornell University under Eva Tardos in 2010. His work lies on the intersection of game theory, learning theory, dynamical systems, and algorithms. Before joining SUTD, Georgios was the Wally Baer and Jeri Weiss postdoctoral scholar in Computing and Mathematical Sciences at Caltech and a Berkeley/Simons Fellow.  During his Georgia Tech postdoc at the Electrical and Computer Engineering department, he was involved in the DARPA Physical Intelligence program which explored and prototyped novel, neuron-inspired computer architectures. Finally, he has held visiting positions at the Center for Information and Computation (CWI/Amsterdam) and the economics departments of Oxford and Johns Hopkins University.


  • PhD in Computer Science, Cornell (2010)
  • MSc in Computer Science, Cornell  (2008)
  • MSc in Applied Mathematics and Logic, National University of Athens (2005)
  • Diploma in Electrical Engineering and Computer Science, National Technical University of Athens (2004)

Research Interests

My main research interests lie in the areas of algorithmic game theory, computational learning theory, multi-agent learning and dynamical systems. I am interested in exploring dynamic phenomena that arise from the interaction of numerous adapting agents. Such systems can either be artificial (e.g., large scale computer networks) or natural (e.g., co-evolving species, socioeconomic networks).

Despite their different origins these systems pose similar analytical challenges. These include computational hardness considerations (due to their large state-spaces), topological complexity issues (due to the possible non-equilibrium limit behaviors), as well as instability issues. Addressing these issues requires a combination of tools from computer science, control theory and topology of dynamical systems. These techniques hold the promise of developing a mathematical language for building flexible, robust, large scale network systems with numerous biological as well as technological applications.

Selected Publications

  1. R. Mehta, G. Piliouras, and I. Panageas. Natural Selection as an Inhibitor of Genetic Diversity: Multiplicative Weights Updates Algorithm and a Conjecture of Haploid Genetics. Innovations in Theoretical Computer Science (ITCS), 2015.
  2. G. Piliouras, C. Nieto-Granda, H.I. Christensen, and J.S. Shamma. Persistent Patterns: Multi-Agent Learning beyond Equilibrium and Utility International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2014.
  3. M.-F. Balcan, S. Krehbiel, G. Piliouras, and J. Shin. Near optimality in covering games by exposing global information ACM Transactions of Economics and Computation, 2014.
  4. G. Piliouras and J.S. Shamma. Optimization Despite Chaos: Convex Relaxations to Complex Limit Sets via Poincaré Recurrence. ACM-SIAM Symposium on Discrete Algorithms (SODA), 2014.
  5. G. Piliouras, E. Nikolova, and J.S. Shamma. Risk Sensitivity of Price of Anarchy under Uncertainty. ACM Conference on Electronic Commerce (EC), 2013.
  6. R.D. Kleinberg, K. Ligett, G. Piliouras, and É. Tardos. Beyond the Nash equilibrium barrier. Symposium on Innovations in Computer Science (ICS), 2011.
  7. R.D. Kleinberg, G. Piliouras, and É. Tardos. Multiplicative updates outperform generic no-regret learning in congestion games. ACM Symposium on Theory of Computing (STOC), 2009.


  • Simons Fellowship (Berkeley/Simons)
  • Wally Baer and Jeri Weiss postdoctoral scholar (Caltech)
  • Olin Fellowship (Cornell)