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 2006.12690cs.LGmath.OCstat.ML
收藏

贝叶斯EM算法收敛性的动力系统方法

A Dynamical Systems Approach for Convergence of the Bayesian EM Algorithm

University of Pennsylvania(宾夕法尼亚大学);MIT-IBM Watson AI Lab, IBM Research(IBM研究院麻省理工学院-IBM沃森人工智能实验室);IBM Thomas J. Watson Research Center, IBM Research(IBM研究院托马斯·J·沃森研究中心);Delft University of Technology(代尔夫特理工大学)

Orlando Romero, Subhro Das, Pin-Yu Chen, Sérgio Pequito

摘要:

在基于系统与控制(S\&C)的优化算法分析的最新进展中,没有足够的工作专门致力于机器学习(ML)算法及其应用。本文解决了这一差距,说明(离散时间)李雅普诺夫稳定性理论可以作为一个强大的工具,以帮助,甚至导致,在分析(和潜在的设计)的优化算法,不一定是基于梯度。本文关注的特定ML问题是通过称为最大后验期望最大化(MAP-EM)的流行优化算法在不完全数据贝叶斯框架中进行参数估计。根据动力系统稳定性理论的第一性原理,给出了MAP-EM收敛的条件。此外,如果满足额外的假设,我们表明,快速收敛(线性或二次)实现,这可能是很难揭开没有我们采用的S\&C方法。本文中的收敛保证有效地扩展了EM应用的充分条件集,从而展示了其他ML算法的类似基于S\& C的收敛分析的潜力。

英文摘要:

Out of the recent advances in systems and control (S\&C)-based analysis of optimization algorithms, not enough work has been specifically dedicated to machine learning (ML) algorithms and its applications. This paper addresses this gap by illustrating how (discrete-time) Lyapunov stability theory can serve as a powerful tool to aid, or even lead, in the analysis (and potential design) of optimization algorithms that are not necessarily gradient-based. The particular ML problem that this paper focuses on is that of parameter estimation in an incomplete-data Bayesian framework via the popular optimization algorithm known as maximum a posteriori expectation-maximization (MAP-EM). Following first principles from dynamical systems stability theory, conditions for convergence of MAP-EM are developed. Furthermore, if additional assumptions are met, we show that fast convergence (linear or quadratic) is achieved, which could have been difficult to unveil without our adopted S\&C approach. The convergence guarantees in this paper effectively expand the set of sufficient conditions for EM applications, thereby demonstrating the potential of similar S\&C-based convergence analysis of other ML algorithms.

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。

↑