arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.12649math.OCcs.GTcs.LGcs.MA
收藏

关于多损失Optimization算法整体收敛的不可能性

On the Impossibility of Global Convergence in Multi-Loss Optimization

Alistair Letcher

摘要:

在弱正则性条件下,基于梯度的方法在单损失情况下全局收敛到一个临界点。这是已知的打破香草梯度下降时,移动到多损失优化,但我们可以希望建立一些算法与全球保证?我们消极地解决这个开放的问题,证明理想的收敛性能不能同时持有任何算法。我们的结果更多的是与没有令人满意的结果的游戏的存在,而不是与算法本身。更明确地说,我们构建了一个两个玩家的游戏与零和的相互作用,其损失是强制性的和分析,但其唯一的同时临界点是一个严格的最大值。任何“合理的”算法,定义为避免严格的最大值,因此将无法收敛。这与单一损失有着根本的不同,单一损失中的波动性意味着存在一个全局极小值。此外,我们证明了广泛的现有的基于梯度的方法几乎肯定有界的,但非收敛的迭代在一个适当的小学习率构造零和对策。尽管如此,这种行为是否会出现在ML从业者感兴趣的高维游戏中仍然是一个悬而未决的问题,例如GANs或多智能体RL。

英文摘要:

Under mild regularity conditions, gradient-based methods converge globally to a critical point in the single-loss setting. This is known to break down for vanilla gradient descent when moving to multi-loss optimization, but can we hope to build some algorithm with global guarantees? We negatively resolve this open problem by proving that desirable convergence properties cannot simultaneously hold for any algorithm. Our result has more to do with the existence of games with no satisfactory outcomes, than with algorithms per se. More explicitly we construct a two-player game with zero-sum interactions whose losses are both coercive and analytic, but whose only simultaneous critical point is a strict maximum. Any 'reasonable' algorithm, defined to avoid strict maxima, will therefore fail to converge. This is fundamentally different from single losses, where coercivity implies existence of a global minimum. Moreover, we prove that a wide range of existing gradient-based methods almost surely have bounded but non-convergent iterates in a constructed zero-sum game for suitably small learning rates. It nonetheless remains an open question whether such behavior can arise in high-dimensional games of interest to ML practitioners, such as GANs or multi-agent RL.

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。

↑