Graph-based regularization for regression problems with alignment and highly-correlated designs.

Sparse models for high-dimensional linear regression and machine learning have received substantial attention over the past two decades. Model selection, or determining which features or covariates are the best explanatory variables, is critical to the interpretability of a learned model. Much of the current literature assumes that covariates are only mildly correlated. However, in many modern applications covariates are highly correlated and do not exhibit key properties (such as the restricted eigenvalue condition, restricted isometry property, or other related assumptions). This work considers a high-dimensional regression setting in which a graph governs both correlations among the covariates and the similarity among regression coefficients - meaning there is alignment between the covariates and regression coefficients. Using side information about the strength of correlations among features, we form a graph with edge weights corresponding to pairwise covariances. This graph is used to define a graph total variation regularizer that promotes similar weights for correlated features. This work shows how the proposed graph-based regularization yields mean-squared error guarantees for a broad range of covariance graph structures. These guarantees are optimal for many specific covariance graphs, including block and lattice graphs. Our proposed approach outperforms other methods for highly-correlated design in a variety of experiments on synthetic data and real biochemistry data.

A Phase I Dose Escalation Study Demonstrates Quercetin Safety and Explores Potential for Bioflavonoid Antivirals in Patients with Chronic Hepatitis C.

The hepatitis C virus (HCV) infects more than 180 million people worldwide, with long-term consequences including liver failure and hepatocellular carcinoma. Quercetin bioflavonoids can decrease HCV production in tissue culture, in part through inhibition of heat shock proteins. If quercetin demonstrates safety and antiviral activity in patients, then it could be developed into an inexpensive HCV treatment for third world countries or other affected populations that lack financial means to cover the cost of mainstream antivirals. A phase 1 dose escalation study was performed to evaluate the safety of quercetin in 30 untreated patients with chronic HCV infection and to preliminarily characterize quercetin's potential in suppressing viral load and/or liver injury. Quercetin displayed safety in all trial participants. Additionally, 8 patients showed a "clinically meaningful" 0.41-log viral load decrease. There was a positive correlation (r = 0.41, p = 0.03) indicating a tendency for HCV decrease in patients with a lower ratio of plasma quercetin relative to dose. No significant changes in aspartate transaminase and alanine transaminase were detected. In conclusion, quercetin exhibited safety (up to 5 g daily) and there was a potential for antiviral activity in some hepatitis C patients.

The Affordable Care Act in brief: a look into January 2014 changes.

The Affordable Care Act (ACA) introduces many new mechanisms into the insurance marketplace that significantly impact business as usual. Although the changes are of great magnitude and, at times, intertwined with each other, this article sets out to introduce many of the changes in a simplified format. The 2014 changes will impact each market (individual, small-group, large-group and self-insured) in a different manner. This overview highlights those differences and includes a discussion of each.

Hyperentanglement source by intersubband two-photon emission from semiconductor quantum wells.

We propose an efficient hyperentanglement source emitting photon pairs entangled in both energy and polarization. The compact electrically driven room-temperature source, based on intersubband two-photon emission from semiconductor quantum wells (QWs) exhibits pair generation rates several orders of magnitude higher than alternative conventional schemes. A theoretical formalism is derived for the calculation of photon pair generation spectra and rates. The results are presented for superlattice structures similar to quantum cascade lasers of GaAs/AlGaAs QWs emitting in the mid-IR and far-IR and for InN/AlN QW structures suitable for telecommunication wavelengths.

Plasma-induced formation of Ag nanodots for ultra-high-enhancement surface-enhanced Raman scattering substrates.

We report here plasma-induced formation of Ag nanostructures for surface-enhanced Raman scattering (SERS) applications. An array of uniform Ag patterned structures of 150 nm diameter was first fabricated on a silicon substrate with imprint lithography; then the substrate was further treated with an oxygen plasma to fracture the patterned structures into clusters of smaller, interconnected, closely packed Ag nanoparticles (20-60 nm) and redeposited Ag nanodots ( approximately 10 nm) between the clusters. The substrate thus formed had a uniform ultrahigh SERS enhancement factor (1010) over the entire substrate for 4-mercaptophenol molecules. By comparison, Au patterned structures fabricated with the same method did not undergo such a morphological change after the plasma treatment and showed no enhancement of Raman scattering.