arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1806.05151cs.LGmath.OCstat.ML
收藏

约束非凸优化的Lagrange函数景观与随机搜索

On Landscape of Lagrangian Functions and Stochastic Search for Constrained Nonconvex Optimization

Georgia Institute of Technology(佐治亚理工学院);University of Minnesota(明尼苏达大学);Princeton University(普林斯顿大学)

Zhehui Chen, Xingguo Li, Lin F. Yang, Jarvis Haupt, Tuo Zhao

摘要:

我们研究机器学习、信号处理和随机控制中的约束非凸优化问题。这是众所周知的,这些问题可以改写为一个极大极小问题的拉格朗日形式。然而,由于缺乏凸性,他们的景观没有得到很好的理解,如何找到稳定的平衡的拉格朗日函数仍然是未知的。为了弥补这一差距,我们研究了拉格朗日函数的景观。此外,我们定义了一类特殊的拉格朗日函数。他们有两个特点:1.平衡点是稳定的或不稳定的(形式定义见第二节); 2.稳定平衡点对应于原问题的全局最优解。我们表明,广义特征值(GEV)问题,包括典型相关分析和其他问题,属于类。具体来说,我们通过利用不变群和对称属性来描述其稳定和不稳定平衡点(更多细节见第3节)。受这些整洁的几何结构,我们提出了一个简单,高效,随机的原始-对偶算法解决在线GEV问题。在理论上,我们给出了充分条件,在此基础上,我们利用扩散近似建立了在线GEV问题的渐近收敛速度,并得到了第一个样本复杂性结果,这在应用概率和随机控制中有着广泛的应用.数值结果支持我们的理论。

英文摘要:

We study constrained nonconvex optimization problems in machine learning, signal processing, and stochastic control. It is well-known that these problems can be rewritten to a minimax problem in a Lagrangian form. However, due to the lack of convexity, their landscape is not well understood and how to find the stable equilibria of the Lagrangian function is still unknown. To bridge the gap, we study the landscape of the Lagrangian function. Further, we define a special class of Lagrangian functions. They enjoy two properties: 1.Equilibria are either stable or unstable (Formal definition in Section 2); 2.Stable equilibria correspond to the global optima of the original problem. We show that a generalized eigenvalue (GEV) problem, including canonical correlation analysis and other problems, belongs to the class. Specifically, we characterize its stable and unstable equilibria by leveraging an invariant group and symmetric property (more details in Section 3). Motivated by these neat geometric structures, we propose a simple, efficient, and stochastic primal-dual algorithm solving the online GEV problem. Theoretically, we provide sufficient conditions, based on which we establish an asymptotic convergence rate and obtain the first sample complexity result for the online GEV problem by diffusion approximations, which are widely used in applied probability and stochastic control. Numerical results are provided to support our theory.

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。

↑