Corpus ID: 235694706

On the Convergence of Stochastic Extragradient for Bilinear Games with Restarted Iteration Averaging

@article{Li2021OnTC,
  title={On the Convergence of Stochastic Extragradient for Bilinear Games with Restarted Iteration Averaging},
  author={Chris Junchi Li and Yao-Liang Yu and Nicolas Loizou and Gauthier Gidel and Yi Ma and Nicolas Le Roux and Michael I. Jordan},
  journal={ArXiv},
  year={2021},
  volume={abs/2107.00464}
}
We study the stochastic bilinear minimax optimization problem, presenting an analysis of the Stochastic ExtraGradient (SEG) method with constant step size, and presenting variations of the method that yield favorable convergence. We first note that the last iterate of the basic SEG method only contracts to a fixed neighborhood of the Nash equilibrium, independent of the step size. This contrasts sharply with the standard setting of minimization where standard stochastic algorithms converge to a… Expand

Figures from this paper

References

SHOWING 1-10 OF 40 REFERENCES
Stochastic Hamiltonian Gradient Methods for Smooth Games
TLDR
This work proposes a novel unbiased estimator for the stochastic Hamiltonian gradient descent (SHGD) and shows that SHGD converges linearly to the neighbourhood of a stationary point, and provides the first global non-asymptotic last-iterate convergence guarantees for certain classes of Stochastic smooth games. Expand
On the convergence of the Stochastic Heavy Ball Method
TLDR
This work provides a comprehensive analysis of the Stochastic Heavy Ball (SHB) method, including a convergence of the last iteration of SHB, establishing a faster rate of convergence than existing bounds on the last iterate of Stochastics Gradient Descent (SGD) in the convex setting. Expand
A Tight and Unified Analysis of Gradient-Based Methods for a Whole Spectrum of Differentiable Games
TLDR
A tight analysis of EG’s convergence rate in games shows that, unlike in convex minimization, EG may be much faster than GD, and it is proved that EG achieves the optimal rate for a wide class of algorithms with any number of extrapolations. Expand
Momentum and Stochastic Momentum for Stochastic Gradient, Newton, Proximal Point and Subspace Descent Methods
TLDR
A novel concept, which is called stochastic momentum, aimed at decreasing the cost of performing the momentum step is proposed, and it is proved that in some sparse data regimes and for sufficiently small momentum parameters, these methods enjoy better overall complexity than methods with deterministic momentum. Expand
Solving variational inequalities with Stochastic Mirror-Prox algorithm
In this paper we consider iterative methods for stochastic variational inequalities (s.v.i.) with monotone operators. Our basic assumption is that the operator possesses both smooth and nonsmoothExpand
On lower iteration complexity bounds for the convex concave saddle point problems
In this paper, we study the lower iteration complexity bounds for finding the saddle point of a strongly convex and strongly concave saddle point problem: $\min_x\max_yF(x,y)$. We restrict theExpand
The Limit Points of (Optimistic) Gradient Descent in Min-Max Optimization
TLDR
This work characterize the limit points of two basic first order methods, namely Gradient Descent/Ascent (GDA) and Optimistic Gradients Descent Ascent (OGDA), and shows that both dynamics avoid unstable critical points for almost all initializations. Expand
On the Adaptivity of Stochastic Gradient-Based Optimization
TLDR
The stochastically controlled stochastic gradient method for composite convex finite-sum optimization problems is presented and it is shown that SCSG is adaptive to both strong convexity and target accuracy. Expand
Accelerating Smooth Games by Manipulating Spectral Shapes
TLDR
This work identifies a continuum between convex minimization, where acceleration is possible using Polyak's momentum, and the worst case where gradient descent is optimal, and proposes an optimal algorithm for bilinear games. Expand
Robust Stochastic Approximation Approach to Stochastic Programming
TLDR
It is intended to demonstrate that a properly modified SA approach can be competitive and even significantly outperform the SAA method for a certain class of convex stochastic problems. Expand
...
1
2
3
4
...