arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1804.06291cs.CV
收藏

稀疏子空间聚类的高效求解器

Efficient Solvers for Sparse Subspace Clustering

University of Massachusetts Lowell(马萨诸塞大学洛厄尔分校);University of Colorado Boulder(科罗拉多大学博尔德分校)

Farhad Pourkamali-Anaraki, James Folberth, Stephen Becker

AI总结:

本文提出基于近端梯度框架的稀疏子空间聚类高效求解算法,统一支持 $\ell_1$ 和 $\ell_0$ 正则化及仿射约束,其中 $\ell_1$ 求解仅需 $O(n^2)$ 计算量,$\ell_0$ 方法首次支持仿射约束且保持低内存,实验显示其对参数和噪声更鲁棒。

摘要:

稀疏子空间聚类(SSC)对位于低维子空间并集附近的 $n$ 个点进行聚类。SSC 模型使用 $\ell_1$ 或 $\ell_0$ 正则化,将每个点表示为其他点的线性或仿射组合。使用 $\ell_1$ 正则化会得到一个凸问题,但需要 $O(n^2)$ 的存储空间,并且通常通过交替方向乘子法求解,其计算量为 $O(n^3)$ 次浮点运算。$\ell_0$ 模型是非凸的,但只需要与 $n$ 成线性关系的内存,并通过正交匹配追踪求解,且无法处理仿射子空间的情形。本文表明,近端梯度框架可以求解 SSC,同时覆盖 $\ell_1$ 和 $\ell_0$ 模型,以及线性和仿射约束。对于 $\ell_1$ 和 $\ell_0$,在存在仿射约束的情况下计算邻近算子的算法此前尚未在 SSC 文献中提出,因此我们推导出一个精确且高效的算法,仅用 $O(n^2)$ 次浮点运算即可求解 $\ell_1$ 情形。在 $\ell_0$ 情形下,我们的算法保持了低内存开销,并且是第一个求解带仿射约束的 SSC-$\ell_0$ 模型的算法。实验表明,我们的算法不依赖敏感的正则化参数,并且对稀疏性错误设定和高噪声的敏感度更低。

英文摘要:

Sparse subspace clustering (SSC) clusters $n$ points that lie near a union of low-dimensional subspaces. The SSC model expresses each point as a linear or affine combination of the other points, using either $\ell_1$ or $\ell_0$ regularization. Using $\ell_1$ regularization results in a convex problem but requires $O(n^2)$ storage, and is typically solved by the alternating direction method of multipliers which takes $O(n^3)$ flops. The $\ell_0$ model is non-convex but only needs memory linear in $n$, and is solved via orthogonal matching pursuit and cannot handle the case of affine subspaces. This paper shows that a proximal gradient framework can solve SSC, covering both $\ell_1$ and $\ell_0$ models, and both linear and affine constraints. For both $\ell_1$ and $\ell_0$, algorithms to compute the proximity operator in the presence of affine constraints have not been presented in the SSC literature, so we derive an exact and efficient algorithm that solves the $\ell_1$ case with just $O(n^2)$ flops. In the $\ell_0$ case, our algorithm retains the low-memory overhead, and is the first algorithm to solve the SSC-$\ell_0$ model with affine constraints. Experiments show our algorithms do not rely on sensitive regularization parameters, and they are less sensitive to sparsity misspecification and high noise.

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。

↑