PCA meets RG.

PCA meets RG.
复制标题

DOI:
10.1007/s10955-017-1770-6
复制
发表时间:
2017-05
影响因子:
1.6
通讯作者:
Bialek W
Bialek W
中科院分区:
物理与天体物理3区
文献类型:
--
作者:
Bradde S;Bialek W

文献摘要

参考文献

被引文献

相似文献

具有多个自由度的系统可以用协方差矩阵来表征;主成分分析(PCA)关注这个矩阵的特征值,希望找到一个更低维的描述。但是,当光谱几乎是连续的时,我们保留的成分和我们忽略的成分之间的任何区别都变得任意了;于是很自然地会问,当我们改变这个任意的截止值时,会发生什么。我们认为,这个问题是类似的动量壳重整化群(RG)。根据这个类比,我们可以定义相关和不相关的算子,其中维数的作用是由特征值密度的性质发挥的。这些结果也提出了一种方法来分析真实的数据。作为一个例子,我们研究了脊椎动物视网膜的神经活动,因为它响应自然主义的电影,并找到行为控制的证据,一个非平凡的固定点。应用于金融数据,我们的分析分离模式占主导地位的采样噪声从一个较小的,但仍然宏观数量的模式描述的非高斯分布。
A system with many degrees of freedom can be characterized by a covariance matrix; principal components analysis (PCA) focuses on the eigenvalues of this matrix, hoping to find a lower dimensional description. But when the spectrum is nearly continuous, any distinction between components that we keep and those that we ignore becomes arbitrary; it then is natural to ask what happens as we vary this arbitrary cutoff. We argue that this problem is analogous to the momentum shell renormalization group (RG). Following this analogy, we can define relevant and irrelevant operators, where the role of dimensionality is played by properties of the eigenvalue density. These results also suggest an approach to the analysis of real data. As an example, we study neural activity in the vertebrate retina as it responds to naturalistic movies, and find evidence of behavior controlled by a nontrivial fixed point. Applied to financial data, our analysis separates modes dominated by sampling noise from a smaller but still macroscopic number of modes described by a non–Gaussian distribution.
DOI: 10.1126/science.1238723
发表时间: 2013-11-01
期刊: SCIENCE
影响因子: 56.9
作者:
Machta, Benjamin B.;Chachra, Ricky;Sethna, James P.
通讯作者: Sethna, James P.
DOI: 10.1007/bf01293604
发表时间: 1981-01-01
期刊: ZEITSCHRIFT FUR PHYSIK B-CONDENSED MATTER
影响因子: --
作者:
BINDER, K
通讯作者: BINDER, K
DOI: 10.1023/a:1004678222262
发表时间: 1999-11-01
影响因子: 1.6
作者:
Coppersmith, SN;Blank, RD;Kadanoff, LP
通讯作者: Kadanoff, LP
DOI: 10.1209/0295-5075/95/47014
发表时间: 2011-08-01
期刊: EPL
影响因子: 1.8
作者:
Castellana, M.
通讯作者: Castellana, M.
DOI: 10.1088/1478-3967/1/3/006
发表时间: 2004-09-01
期刊: PHYSICAL BIOLOGY
影响因子: 2
作者:
Brown, KS;Hill, CC;Cerione, RA
通讯作者: Cerione, RA