Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Georgios Piliouras

Georgios Piliouras

Constants of Motion: The Antidote to Chaos in Optimization and Game Dynamics


Sep 10, 2021
Georgios Piliouras, Xiao Wang

Figure 1 for Constants of Motion: The Antidote to Chaos in Optimization and Game Dynamics
Figure 2 for Constants of Motion: The Antidote to Chaos in Optimization and Game Dynamics
Figure 3 for Constants of Motion: The Antidote to Chaos in Optimization and Game Dynamics
Figure 4 for Constants of Motion: The Antidote to Chaos in Optimization and Game Dynamics

  Access Paper or Ask Questions

Evolutionary Dynamics and $Φ$-Regret Minimization in Games


Jun 28, 2021
Georgios Piliouras, Mark Rowland, Shayegan Omidshafiei, Romuald Elie, Daniel Hennes, Jerome Connor, Karl Tuyls

Figure 1 for Evolutionary Dynamics and $Φ$-Regret Minimization in Games
Figure 2 for Evolutionary Dynamics and $Φ$-Regret Minimization in Games
Figure 3 for Evolutionary Dynamics and $Φ$-Regret Minimization in Games
Figure 4 for Evolutionary Dynamics and $Φ$-Regret Minimization in Games

  Access Paper or Ask Questions

Exploration-Exploitation in Multi-Agent Competition: Convergence with Bounded Rationality


Jun 24, 2021
Stefanos Leonardos, Georgios Piliouras, Kelly Spendlove

Figure 1 for Exploration-Exploitation in Multi-Agent Competition: Convergence with Bounded Rationality
Figure 2 for Exploration-Exploitation in Multi-Agent Competition: Convergence with Bounded Rationality
Figure 3 for Exploration-Exploitation in Multi-Agent Competition: Convergence with Bounded Rationality
Figure 4 for Exploration-Exploitation in Multi-Agent Competition: Convergence with Bounded Rationality

  Access Paper or Ask Questions

Online Optimization in Games via Control Theory: Connecting Regret, Passivity and Poincaré Recurrence


Jun 15, 2021
Yun Kuen Cheung, Georgios Piliouras

Figure 1 for Online Optimization in Games via Control Theory: Connecting Regret, Passivity and Poincaré Recurrence
Figure 2 for Online Optimization in Games via Control Theory: Connecting Regret, Passivity and Poincaré Recurrence
Figure 3 for Online Optimization in Games via Control Theory: Connecting Regret, Passivity and Poincaré Recurrence
Figure 4 for Online Optimization in Games via Control Theory: Connecting Regret, Passivity and Poincaré Recurrence
* In ICML 2021 
  Access Paper or Ask Questions

Efficient Online Learning for Dynamic k-Clustering


Jun 08, 2021
Dimitris Fotakis, Georgios Piliouras, Stratis Skoulakis

Figure 1 for Efficient Online Learning for Dynamic k-Clustering
Figure 2 for Efficient Online Learning for Dynamic k-Clustering
Figure 3 for Efficient Online Learning for Dynamic k-Clustering
Figure 4 for Efficient Online Learning for Dynamic k-Clustering

  Access Paper or Ask Questions

Global Convergence of Multi-Agent Policy Gradient in Markov Potential Games


Jun 03, 2021
Stefanos Leonardos, Will Overman, Ioannis Panageas, Georgios Piliouras

Figure 1 for Global Convergence of Multi-Agent Policy Gradient in Markov Potential Games
Figure 2 for Global Convergence of Multi-Agent Policy Gradient in Markov Potential Games
Figure 3 for Global Convergence of Multi-Agent Policy Gradient in Markov Potential Games
Figure 4 for Global Convergence of Multi-Agent Policy Gradient in Markov Potential Games

  Access Paper or Ask Questions

Learning in Matrix Games can be Arbitrarily Complex


Mar 05, 2021
Gabriel P. Andrade, Rafael Frongillo, Georgios Piliouras

Figure 1 for Learning in Matrix Games can be Arbitrarily Complex
Figure 2 for Learning in Matrix Games can be Arbitrarily Complex
Figure 3 for Learning in Matrix Games can be Arbitrarily Complex

  Access Paper or Ask Questions

Scaling up Mean Field Games with Online Mirror Descent


Feb 28, 2021
Julien Perolat, Sarah Perrin, Romuald Elie, Mathieu Laurière, Georgios Piliouras, Matthieu Geist, Karl Tuyls, Olivier Pietquin

Figure 1 for Scaling up Mean Field Games with Online Mirror Descent
Figure 2 for Scaling up Mean Field Games with Online Mirror Descent
Figure 3 for Scaling up Mean Field Games with Online Mirror Descent
Figure 4 for Scaling up Mean Field Games with Online Mirror Descent

  Access Paper or Ask Questions

Follow-the-Regularized-Leader Routes to Chaos in Routing Games


Feb 17, 2021
Jakub Bielawski, Thiparat Chotibut, Fryderyk Falniowski, Grzegorz Kosiorowski, Michał Misiurewicz, Georgios Piliouras

Figure 1 for Follow-the-Regularized-Leader Routes to Chaos in Routing Games
Figure 2 for Follow-the-Regularized-Leader Routes to Chaos in Routing Games
Figure 3 for Follow-the-Regularized-Leader Routes to Chaos in Routing Games
Figure 4 for Follow-the-Regularized-Leader Routes to Chaos in Routing Games
* 30 pages, 8 figures 
  Access Paper or Ask Questions

Solving Min-Max Optimization with Hidden Structure via Gradient Descent Ascent


Jan 13, 2021
Lampros Flokas, Emmanouil-Vasileios Vlatakis-Gkaragkounis, Georgios Piliouras

Figure 1 for Solving Min-Max Optimization with Hidden Structure via Gradient Descent Ascent
Figure 2 for Solving Min-Max Optimization with Hidden Structure via Gradient Descent Ascent
Figure 3 for Solving Min-Max Optimization with Hidden Structure via Gradient Descent Ascent
Figure 4 for Solving Min-Max Optimization with Hidden Structure via Gradient Descent Ascent

  Access Paper or Ask Questions

Evolutionary Game Theory Squared: Evolving Agents in Endogenously Evolving Zero-Sum Games


Dec 15, 2020
Stratis Skoulakis, Tanner Fiez, Ryann Sim, Georgios Piliouras, Lillian Ratliff

Figure 1 for Evolutionary Game Theory Squared: Evolving Agents in Endogenously Evolving Zero-Sum Games
Figure 2 for Evolutionary Game Theory Squared: Evolving Agents in Endogenously Evolving Zero-Sum Games
Figure 3 for Evolutionary Game Theory Squared: Evolving Agents in Endogenously Evolving Zero-Sum Games
Figure 4 for Evolutionary Game Theory Squared: Evolving Agents in Endogenously Evolving Zero-Sum Games
* To appear in AAAI 2021 
  Access Paper or Ask Questions

Efficient Online Learning of Optimal Rankings: Dimensionality Reduction via Gradient Descent


Nov 05, 2020
Dimitris Fotakis, Thanasis Lianeas, Georgios Piliouras, Stratis Skoulakis

Figure 1 for Efficient Online Learning of Optimal Rankings: Dimensionality Reduction via Gradient Descent

  Access Paper or Ask Questions

No-regret learning and mixed Nash equilibria: They do not mix


Oct 20, 2020
Lampros Flokas, Emmanouil-Vasileios Vlatakis-Gkaragkounis, Thanasis Lianeas, Panayotis Mertikopoulos, Georgios Piliouras

Figure 1 for No-regret learning and mixed Nash equilibria: They do not mix
Figure 2 for No-regret learning and mixed Nash equilibria: They do not mix
Figure 3 for No-regret learning and mixed Nash equilibria: They do not mix
Figure 4 for No-regret learning and mixed Nash equilibria: They do not mix
* 24 pages, 7 figures, 1 table 
  Access Paper or Ask Questions

No-regreet learning and mixed Nash equilibria: They do not mix


Oct 19, 2020
Lampros Flokas, Emmanouil-Vasileios Vlatakis-Gkaragkounis, Thanasis Lianeas, Panayotis Mertikopoulos, Georgios Piliouras

Figure 1 for No-regreet learning and mixed Nash equilibria: They do not mix
Figure 2 for No-regreet learning and mixed Nash equilibria: They do not mix
Figure 3 for No-regreet learning and mixed Nash equilibria: They do not mix
Figure 4 for No-regreet learning and mixed Nash equilibria: They do not mix
* 24 pages, 7 figures, 1 table 
  Access Paper or Ask Questions

Chaos, Extremism and Optimism: Volume Analysis of Learning in Games


May 28, 2020
Yun Kuen Cheung, Georgios Piliouras

Figure 1 for Chaos, Extremism and Optimism: Volume Analysis of Learning in Games
Figure 2 for Chaos, Extremism and Optimism: Volume Analysis of Learning in Games
Figure 3 for Chaos, Extremism and Optimism: Volume Analysis of Learning in Games
Figure 4 for Chaos, Extremism and Optimism: Volume Analysis of Learning in Games
* 20 pages, 4 figures 
  Access Paper or Ask Questions

From Poincaré Recurrence to Convergence in Imperfect Information Games: Finding Equilibrium via Regularization


Feb 19, 2020
Julien Perolat, Remi Munos, Jean-Baptiste Lespiau, Shayegan Omidshafiei, Mark Rowland, Pedro Ortega, Neil Burch, Thomas Anthony, David Balduzzi, Bart De Vylder, Georgios Piliouras, Marc Lanctot, Karl Tuyls

Figure 1 for From Poincaré Recurrence to Convergence in Imperfect Information Games: Finding Equilibrium via Regularization
Figure 2 for From Poincaré Recurrence to Convergence in Imperfect Information Games: Finding Equilibrium via Regularization
Figure 3 for From Poincaré Recurrence to Convergence in Imperfect Information Games: Finding Equilibrium via Regularization
Figure 4 for From Poincaré Recurrence to Convergence in Imperfect Information Games: Finding Equilibrium via Regularization
* 43 pages 
  Access Paper or Ask Questions

Smooth markets: A basic mechanism for organizing gradient-based learners


Jan 18, 2020
David Balduzzi, Wojciech M Czarnecki, Thomas W Anthony, Ian M Gemp, Edward Hughes, Joel Z Leibo, Georgios Piliouras, Thore Graepel

Figure 1 for Smooth markets: A basic mechanism for organizing gradient-based learners
Figure 2 for Smooth markets: A basic mechanism for organizing gradient-based learners
Figure 3 for Smooth markets: A basic mechanism for organizing gradient-based learners
* ICLR 2020 
 * 18 pages, 3 figures 
  Access Paper or Ask Questions

Multiagent Evaluation under Incomplete Information


Oct 30, 2019
Mark Rowland, Shayegan Omidshafiei, Karl Tuyls, Julien Perolat, Michal Valko, Georgios Piliouras, Remi Munos

Figure 1 for Multiagent Evaluation under Incomplete Information
Figure 2 for Multiagent Evaluation under Incomplete Information
Figure 3 for Multiagent Evaluation under Incomplete Information
Figure 4 for Multiagent Evaluation under Incomplete Information

  Access Paper or Ask Questions

Efficiently avoiding saddle points with zero order methods: No gradients required


Oct 29, 2019
Lampros Flokas, Emmanouil-Vasileios Vlatakis-Gkaragkounis, Georgios Piliouras

Figure 1 for Efficiently avoiding saddle points with zero order methods: No gradients required
Figure 2 for Efficiently avoiding saddle points with zero order methods: No gradients required
Figure 3 for Efficiently avoiding saddle points with zero order methods: No gradients required
* To appear in NeurIPS 2019 
  Access Paper or Ask Questions

Poincaré Recurrence, Cycles and Spurious Equilibria in Gradient-Descent-Ascent for Non-Convex Non-Concave Zero-Sum Games


Oct 28, 2019
Lampros Flokas, Emmanouil-Vasileios Vlatakis-Gkaragkounis, Georgios Piliouras

Figure 1 for Poincaré Recurrence, Cycles and Spurious Equilibria in Gradient-Descent-Ascent for Non-Convex Non-Concave Zero-Sum Games
Figure 2 for Poincaré Recurrence, Cycles and Spurious Equilibria in Gradient-Descent-Ascent for Non-Convex Non-Concave Zero-Sum Games
Figure 3 for Poincaré Recurrence, Cycles and Spurious Equilibria in Gradient-Descent-Ascent for Non-Convex Non-Concave Zero-Sum Games
Figure 4 for Poincaré Recurrence, Cycles and Spurious Equilibria in Gradient-Descent-Ascent for Non-Convex Non-Concave Zero-Sum Games
* To appear in NeurIPS 2019 (Spotlight talk) 
  Access Paper or Ask Questions

Fast and Furious Learning in Zero-Sum Games: Vanishing Regret with Non-Vanishing Step Sizes


May 11, 2019
James P. Bailey, Georgios Piliouras

Figure 1 for Fast and Furious Learning in Zero-Sum Games: Vanishing Regret with Non-Vanishing Step Sizes
Figure 2 for Fast and Furious Learning in Zero-Sum Games: Vanishing Regret with Non-Vanishing Step Sizes
Figure 3 for Fast and Furious Learning in Zero-Sum Games: Vanishing Regret with Non-Vanishing Step Sizes
Figure 4 for Fast and Furious Learning in Zero-Sum Games: Vanishing Regret with Non-Vanishing Step Sizes

  Access Paper or Ask Questions

Multi-Agent Learning in Network Zero-Sum Games is a Hamiltonian System


Mar 05, 2019
James P. Bailey, Georgios Piliouras

Figure 1 for Multi-Agent Learning in Network Zero-Sum Games is a Hamiltonian System
Figure 2 for Multi-Agent Learning in Network Zero-Sum Games is a Hamiltonian System

  Access Paper or Ask Questions

Short-distance commuters in the smart city


Feb 16, 2019
Francisco Benita, Garvit Bansal, Georgios Piliouras, Bige Tunçer

Figure 1 for Short-distance commuters in the smart city
Figure 2 for Short-distance commuters in the smart city
Figure 3 for Short-distance commuters in the smart city
Figure 4 for Short-distance commuters in the smart city
* 21 pages 
  Access Paper or Ask Questions

Venn GAN: Discovering Commonalities and Particularities of Multiple Distributions


Feb 09, 2019
Yasin Yazıcı, Bruno Lecouat, Chuan-Sheng Foo, Stefan Winkler, Kim-Hui Yap, Georgios Piliouras, Vijay Chandrasekhar


  Access Paper or Ask Questions

Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile


Oct 01, 2018
Panayotis Mertikopoulos, Bruno Lecouat, Houssam Zenati, Chuan-Sheng Foo, Vijay Chandrasekhar, Georgios Piliouras

Figure 1 for Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
Figure 2 for Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
Figure 3 for Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
Figure 4 for Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
* 26 pages, 14 figures 
  Access Paper or Ask Questions

The Unusual Effectiveness of Averaging in GAN Training


Jun 12, 2018
Yasin Yazıcı, Chuan-Sheng Foo, Stefan Winkler, Kim-Hui Yap, Georgios Piliouras, Vijay Chandrasekhar

Figure 1 for The Unusual Effectiveness of Averaging in GAN Training
Figure 2 for The Unusual Effectiveness of Averaging in GAN Training
Figure 3 for The Unusual Effectiveness of Averaging in GAN Training
Figure 4 for The Unusual Effectiveness of Averaging in GAN Training

  Access Paper or Ask Questions

First-order Methods Almost Always Avoid Saddle Points


Oct 20, 2017
Jason D. Lee, Ioannis Panageas, Georgios Piliouras, Max Simchowitz, Michael I. Jordan, Benjamin Recht

Figure 1 for First-order Methods Almost Always Avoid Saddle Points

  Access Paper or Ask Questions