arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.09396stat.MLcs.ITcs.LGmath.IT
收藏

基于梯度下降与小随机初始化的秩-1矩阵补全

Rank-1 Matrix Completion with Gradient Descent and Small Random Initialization

Samsung Electronics(三星电子);KAIST(韩国科学技术院)

Daesung Kim, Hye Won Chung

AI总结:

本文研究秩-1对称矩阵补全问题,证明梯度下降在小随机初始化下可收敛到真实解,并给出初始化规模上界及隐式正则化在收敛过程中的关键作用。

摘要:

近年来,矩阵补全问题的非凸形式因其相较于凸形式具有可承受的计算复杂度而受到广泛关注。梯度下降(GD)是求解非凸优化问题的一种简单而高效的基线算法。在理论与实践中的许多不同问题上,当GD与随机初始化相结合时,其成功已经得到验证。然而,以往关于矩阵补全的工作要么需要精心设计的初始化,要么需要正则化项来证明GD的收敛性。本文研究秩-1对称矩阵补全问题,并证明当使用小随机初始化时,GD能够收敛到真实值。我们证明,在对数数量级的迭代次数内,轨迹会进入发生局部收敛的区域。我们给出了足以保证收敛的初始化规模的上界,并表明当可用样本更多时,可以使用更大的初始化。我们观察到,GD的隐式正则化效应在分析中起着关键作用,并且对于整个轨迹而言,它能够防止每个元素变得比其他元素大得多。

英文摘要:

The nonconvex formulation of the matrix completion problem has received significant attention in recent years due to its affordable complexity compared to the convex formulation. Gradient Descent (GD) is a simple yet efficient baseline algorithm for solving nonconvex optimization problems. The success of GD has been witnessed in many different problems in both theory and practice when it is combined with random initialization. However, previous works on matrix completion require either careful initialization or regularizers to prove the convergence of GD. In this paper, we study the rank-1 symmetric matrix completion and prove that GD converges to the ground truth when small random initialization is used. We show that in a logarithmic number of iterations, the trajectory enters the region where local convergence occurs. We provide an upper bound on the initialization size that is sufficient to guarantee the convergence, and show that a larger initialization can be used as more samples are available. We observe that the implicit regularization effect of GD plays a critical role in the analysis, and for the entire trajectory, it prevents each entry from becoming much larger than the others.

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。

↑