We showcase the versatility of the framework by designing novel proximal algorithms for state-of-the-art regression and variable selection schemes in high-dimensional statistics.
Sparsity is one of the most important notions in recent signal/image processing [1], machine learning [2], communications engineering [3], and high-dimensional statistics [4].
Under mild assumptions, we analyse the mean square error (MSE) of the estimators of one-dimensional statistics of the optimal filtering distribution and show explicitly the effect of the parallelisation scheme on the convergence rate.
In addition to providing new insights on the physical nature of the dissipation fields, the two-dimensional statistics of the mixture fraction and scalar dissipation rate are examined in detail, in the form of conditional averaging and through the construction of probability density functions.
The morphological group of features is traditionally used in clinical practice and it mainly summaries the one-dimensional statistics.
In a nutshell, this paper presents measures of association for systems of genetic variables that go beyond standard two-dimensional statistics.
Building on the multi-task regularized-regression framework recently studied in machine learning and high-dimensional statistics (Meier et al., 2008; Obozinski et al., 2008; Yuan and Lin, 2006; Zhao et al., 2008), we describe our MPGL algorithm using L1/ L2-regularization that can maximize the power for detecting SNPs that affect more than one population as well as population-specific causal SNPs.
It turns out that the modular structure can be exploited in such a situation, to split a high-dimensional problem into subproblems involving only lower-dimensional summary statistics.
Hall effect measurements are performed at temperatures between 77 and 300 K. Temperature dependence of carrier mobility is analysed by an analytical formula based on two-dimensional degenerate statistics by taking into account all major scattering mechanisms for a two-dimensional electron gas confined in a triangular quantum well between GaxIn1-xN epilayer and GaN buffer.
Industry executives in the United States said the Insead report was a significant counterweight to the one-dimensional O.E.C.D. statistics.
