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 2202.05352cs.LGcs.CVcs.GT
收藏

领域对抗训练:游戏视角

Domain Adversarial Training: A Game Perspective

NVIDIA(英伟达);University of Toronto(多伦多大学);Vector Institute(向量研究所);University of Waterloo(滑铁卢大学)

David Acuna, Marc T Law, Guojun Zhang, Sanja Fidler

摘要:

领域自适应的主要工作集中在使用领域对抗训练来学习不变表示。在本文中,我们从博弈论的角度来解释这种方法。将领域对抗训练中的最优解定义为局部纳什均衡,我们证明了领域对抗训练中的梯度下降可能违反优化器的渐近收敛保证,通常会阻碍传输性能。我们的分析使我们用高阶ODE求解器(即,Runge-Kutta),我们得到渐近收敛保证。这一系列优化器明显更稳定,并允许更积极的学习率,从而在用作标准优化器的直接替代品时获得更高的性能增益。我们的实验表明,结合最先进的领域对抗方法,我们在不到一半的训练迭代中实现了高达3.5%的改进。我们的优化器很容易实现,没有额外的参数,并且可以插入任何领域对抗框架。

英文摘要:

The dominant line of work in domain adaptation has focused on learning invariant representations using domain-adversarial training. In this paper, we interpret this approach from a game theoretical perspective. Defining optimal solutions in domain-adversarial training as a local Nash equilibrium, we show that gradient descent in domain-adversarial training can violate the asymptotic convergence guarantees of the optimizer, oftentimes hindering the transfer performance. Our analysis leads us to replace gradient descent with high-order ODE solvers (i.e., Runge-Kutta), for which we derive asymptotic convergence guarantees. This family of optimizers is significantly more stable and allows more aggressive learning rates, leading to high performance gains when used as a drop-in replacement over standard optimizers. Our experiments show that in conjunction with state-of-the-art domain-adversarial methods, we achieve up to 3.5% improvement with less than of half training iterations. Our optimizers are easy to implement, free of additional parameters, and can be plugged into any domain-adversarial framework.

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。

↑