arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1806.07268cs.LGcs.GTstat.ML
收藏

对抗网络的超越局部纳什均衡

Beyond Local Nash Equilibria for Adversarial Networks

Frans A. Oliehoek, Rahul Savani, Jose Gallego, Elise van der Pol, Roderich Groß

摘要:

除了一些特殊情况外,目前用于生成对抗网络(GANs)的训练方法最多可以保证收敛到“局部纳什均衡”(LNE)。然而,这样的LNE可以任意远离实际的纳什均衡(NE),这意味着对所找到的生成器或分类器的质量没有保证。本文提出将GANs明确建模为混合策略中的有限游戏,从而确保每个LNE都是NE。有了这个公式,我们提出了一个解决方案的方法,被证明是单调收敛到资源有界的纳什均衡(RB-NE):通过增加计算资源,我们可以找到更好的解决方案。我们的经验表明,我们的方法是不太容易出现典型的GAN问题,如模式坍塌,并产生解决方案,是较少利用比GANs和MGANs,并密切类似于理论预测的NE。

英文摘要:

Save for some special cases, current training methods for Generative Adversarial Networks (GANs) are at best guaranteed to converge to a `local Nash equilibrium` (LNE). Such LNEs, however, can be arbitrarily far from an actual Nash equilibrium (NE), which implies that there are no guarantees on the quality of the found generator or classifier. This paper proposes to model GANs explicitly as finite games in mixed strategies, thereby ensuring that every LNE is an NE. With this formulation, we propose a solution method that is proven to monotonically converge to a resource-bounded Nash equilibrium (RB-NE): by increasing computational resources we can find better solutions. We empirically demonstrate that our method is less prone to typical GAN problems such as mode collapse, and produces solutions that are less exploitable than those produced by GANs and MGANs, and closely resemble theoretical predictions about NEs.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑