3 papers from Criteo AI Lab folks at COLT 21:
- The last-iterate convergence rate of optimistic mirror descent in stochastic variational inequalities, W. Azizian (ENS), F. Iutzeler (UGA), J. Malick (CNRS), and P. Mertikopoulos (Criteo AI Lab/CNRS)
- Survival of the strictest: Stable and unstable equilibria under regularized learning with partial information, A. Giannou (NTUA), E. M. Vlatakis-Gkaragkounis (Columbia), and and P. Mertikopoulos (Criteo AI Lab/CNRS)
- Adaptive learning in continuous games: Optimal regret bounds and convergence to Nash equilibrium, Y. G. Hsieh (UGA), K. Antonakopoulos (UGA), and P. Mertikopoulos (Criteo AI Lab/CNRS)