A Variational Inequality Perspective on Generative Adversarial Nets

By Gauthier Gidel, Hugo Berard, Pascal Vincent, Simon Lacoste-Julien. 2018

32 citations recorded by the source corpus.

Subfields: Generative Models, Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Generative Models

In Probabilistic & Bayesian Methods

Citations in the corpus

A Tight and Unified Analysis of Extragradient for a Whole Spectrum of Differentiable Games (2019, methodology): “Gidel et al. (2019a) interpreted GANs as a variational inequality problem and derive a variant of extragradient which avoids "wasting" a gradient.”
A Closer Look at the Optimization Landscapes of Generative Adversarial Networks (2019, methodology): “This could explain why averaging and extragradient have been shown as promising methods to improve performances (Gidel et al., 2019; Yazıcı et al., 2019).”
Reducing Noise in GAN Training with Variance Reduced Extragradient (2019, methodology): “The gradient method does not converge for some convex-concave examples (Mescheder et al., 2017; Gidel et al., 2019a).”
Negative Momentum for Improved Game Dynamics (2018, methodology): “(2017) and Gidel et al. (2018), the GAN objective has the following form,  θ∗ ∈ arg min θ∈Θ LG(θ,φ) φ∗ ∈ arg min φ∈Φ LD(θ,φ) .”
Which Training Methods for GANs do actually Converge? (2018, methodology): “Similarly to prior work (Karras et al., 2017; Yazici et al., 2018; Gidel et al., 2018), we use an exponential moving average 16 with decay 0 .”
The Unusual Effectiveness of Averaging in GAN Training (2018, methodology): “Averaging at minimax objective: Gidel et al. (2018) used uniform averaging similar to our work, however we show its shortcomings and how exponential decay averaging performs better.”
A Unified Analysis of Extra-gradient and Optimistic Gradient Methods for Saddle Point Problems: Proximal Point Approach (2019, methodology): “…complexity of O ( κ 2 log(1 /ǫ )) for EG, without anlyzing the general strongly convex-strongly concave setting. another In recent independent work, Gidel et al. (2018) analyzed the convergence of the OGDA method using the interpretation that OGDA is a variant of EG using extrapolation from the…”
A Domain Agnostic Measure for Monitoring and Evaluating GANs (2018, background): “Some exceptions are [5, 10, 8, 13] but these works do not address the problem of evaluation.”
Local Saddle Point Optimization: A Curvature Exploitation Approach (2018, methodology): “These oscillations can be addressed by averaging the iterates [32] or using the extragradient method (a perturbed version of the gradient method) [19, 13].”
On Finding Local Nash Equilibria (and Only Local Nash Equilibria) in Zero-Sum Games (2019, unknown): “A second promising line of research into theoretically sound methods of ﬁnding the Nash equilibria of zero-sum games has approached the issue from the perspective of variational inequalities (Mertikopoulos et al., 2018a; Gidel et al., 2018).”