arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1306.5825cs.LGcs.DSstat.ML
收藏

Fourier PCA与鲁棒张量分解

Fourier PCA and Robust Tensor Decomposition

Microsoft Research India(微软研究院印度);Georgia Tech(佐治亚理工学院);CMU(卡内基梅隆大学)

Navin Goyal, Santosh Vempala, Ying Xiao

摘要:

傅立叶主成分分析是一个矩阵的主成分分析,该矩阵是从分布的傅立叶变换的对数的高阶导数获得的。我们通过对在秩- $1$分解中共享相同向量的一对张量开发一种张量分解方法,使这种方法算法化。我们的主要应用是欠定ICA的第一个可证明多项式时间算法,即,从观测值$y=Ax$学习$n \times m$矩阵$A$,其中$x$从具有任意非高斯分量的未知乘积分布中得出。分量分布的数目$m$可以任意大于维数$n$,并且$A$的列只需要满足自然且有效可验证的非简并性条件。作为第二个应用程序,我们给出了一个替代算法学习混合球高斯线性独立的手段。这些结果也适用于高斯噪声的存在。

英文摘要:

Fourier PCA is Principal Component Analysis of a matrix obtained from higher order derivatives of the logarithm of the Fourier transform of a distribution.We make this method algorithmic by developing a tensor decomposition method for a pair of tensors sharing the same vectors in rank-$1$ decompositions. Our main application is the first provably polynomial-time algorithm for underdetermined ICA, i.e., learning an $n \times m$ matrix $A$ from observations $y=Ax$ where $x$ is drawn from an unknown product distribution with arbitrary non-Gaussian components. The number of component distributions $m$ can be arbitrarily higher than the dimension $n$ and the columns of $A$ only need to satisfy a natural and efficiently verifiable nondegeneracy condition. As a second application, we give an alternative algorithm for learning mixtures of spherical Gaussians with linearly independent means. These results also hold in the presence of Gaussian 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。

↑