Is my model any good: differentially private regression diagnostics

Linear and logistic regression are popular statistical techniques for analyzing multi-variate data. Typically, analysts do not simply posit a particular form of the regression model, estimate its parameters, and use the results for inference or prediction. Instead, they first use a variety of diagnostic techniques to assess how well the model fits the relationships in the data and how well it can be expected to predict outcomes for out-of-sample records, revising the model as necessary to improve fit and predictive power. In this article, we develop \(\epsilon \)-differentially private diagnostics tools for regression, beginning to fill a gap in privacy-preserving data analysis. Specifically, we create differentially private versions of residual plots for linear regression and of receiver operating characteristic (ROC) curves as well as binned residual plot for logistic regression. The residual plot and binned residual plot help determine whether or not the data satisfy the assumptions underlying the regression model, and the ROC curve is used to assess the predictive power of the logistic regression model. These diagnostics improve the usefulness of algorithms for computing differentially private regression output, which alone does not allow analysts to assess the quality of the posited model. Our empirical studies show that these algorithms can be effective tools for allowing users to evaluate the quality of their models.     

Impact of growth rate on the quality of ZNS-MQW InGaAsP/InP laser structures grown by LP-MOVPE

We investigated the influence of the growth rate on the quality of zero-net-strained InGaAsP/InGaAsP/InP multiquantum well structures for 1.55 μm emission grown by low pressure metalorganic vapor phase epitaxy. The samples consisted of fixed compressive strained wells (ɛ=+1%) and tensile strained barriers (ɛ=−0.5%) grown with different quaternary bandgap wavelengths (λ_B=1.1–1.4 μm). Using higher growth rates, we obtained for the first time high quality zero net strained multi quantum well structures, regardless having constant group V composition in the well and barriers. The samples were analyzed by x-ray diffraction, photoluminescence and atomic force microscopy techniques. The amplitude of surface modulation roughness along [011] direction decreased from 20 nm to 0.53 nm with increasing growth rate and/or quaternary compositions grown outside the miscibility gap. A new deep PL broad emission band strongly correlated with the onset of wavy layer growth is also reported. Broad area and ridge waveguide lasers with 10 wells exhibited low losses (34 cm⁻¹) and low threshold current densities at infinite cavity length (1020 A·cm⁻² and 1190 A·cm⁻², respectively).     

An ensemble-based system for microaneurysm detection and diabetic retinopathy grading

Reliable microaneurysm detection in digital fundus images is still an open issue in medical image processing. We propose an ensemble-based framework to improve microaneurysm detection. Unlike the well-known approach of considering the output of multiple classifiers, we propose a combination of internal components of microaneurysm detectors, namely preprocessing methods and candidate extractors. We have evaluated our approach for microaneurysm detection in an online competition, where this algorithm is currently ranked as first, and also on two other databases. Since microaneurysm detection is decisive in diabetic retinopathy (DR) grading, we also tested the proposed method for this task on the publicly available Messidor database, where a promising AUC 0.90 ± 0.01 is achieved in a "DR/non-DR"-type classification based on the presence or absence of the microaneurysms.     

Variational optical flow computation in real time.

This paper investigates the usefulness of bidirectional multigrid methods for variational optical flow computations. Although these numerical schemes are among the fastest methods for solving equation systems, they are rarely applied in the field of computer vision. We demonstrate how to employ those numerical methods for the treatment of variational optical flow formulations and show that the efficiency of this approach even allows for real-time performance on standard PCs. As a representative for variational optic flow methods, we consider the recently introduced combined local-global method. It can be considered as a noise-robust generalization of the Horn and Schunck technique. We present a decoupled, as well as a coupled, version of the classical Gauss-Seidel solver, and we develop several multgrid implementations based on a discretization coarse grid approximation. In contrast, with standard bidirectional multigrid algorithms, we take advantage of intergrid transfer operators that allow for nondyadic grid hierarchies. As a consequence, no restrictions concerning the image size or the number of traversed levels have to be imposed. In the experimental section, we juxtapose the developed multigrid schemes and demonstrate their superior performance when compared to unidirectional multgrid methods and nonhierachical solvers. For the well-known 316 x 252 Yosemite sequence, we succeeded in computing the complete set of dense flow fields in three quarters of a second on a 3.06-GHz Pentium4 PC. This corresponds to a frame rate of 18 flow fields per second which outperforms the widely-used Gauss-Seidel method by almost three orders of magnitude.     

An end‐to‐end alternative to TCP PEPs: Initial Spreading,a TCP fast start‐up mechanism

While Transmission Control Protocol (TCP) Performance Enhancing Proxy (PEP) solutions have long been undisputed to solve the inherent satellite problems, the improvement of the regular end‐to‐end TCP congestion avoidance algorithms and the recent emphasis on the PEPs drawbacks have opened the question of the PEPs sustainability. Nevertheless, with a vast majority of Internet connections shorter than ten segments, TCP PEPs continue to be required to counter the poor efficiency of the end‐to‐end TCP start‐up mechanisms. To reduce the PEPs dependency, designing a new fast start‐up TCP mechanism is therefore a major concern. But, while enlarging the Initial Window (IW) up to ten segments is, without any doubt, the fastest solution to deal with a short‐lived connection in an uncongested network, numerous researchers are concerned about the impact of the large initial burst on an already congested network. Based on traffic observations and real experiments, Initial Spreading has been designed to remove those concerns whatever the load and type of networks. It offers performance similar to a large IW in uncongested network and outperforms existing end‐to‐end solutions in congested networks. In this paper, we show that Initial Spreading, taking care of the satellite specificities, is an efficient end‐to‐end alternative to the TCP PEPs. Copyright © 2015 John Wiley & Sons, Ltd.     

The effect of the martensitic packet size on the machinability of modified AISI P20 prehardened mold steel

The effect of martensitic packet size on the machinability of prehardened mold steel at a hardness of ~40 HRC (typical hardness for prehardened mold steels) was studied in terms of cutting force and tool life. The machinability tests were performed in end milling using coated, cemented carbide tools at three different cutting speeds. The results showed that an increase in the martensite packet size led to higher cutting force and shorter tool life. The increase in cutting force was related to the increase of work hardening. The work material with a coarser martensite packet size showed a higher amount of work hardening that can explain the higher cutting force. The longer tool life in the workpieces with finer structure was correlated to smaller amplitude of the variation in cutting force.     

Optimizing location and size of rural schools in Chile

The Chilean Ministry of Education oversees preschool, primary, and secondary education in both urban and rural areas. Many parts of Chile are sparsely populated and there are currently over 4,000 rural schools (almost 38% of all schools in Chile) educating 9.5% of the students in the country. Many of the rural schools are small with only one teacher responsible for instruction of all local students (multigrade schools). The geographical distribution of the rural schools has not been coordinated and this has resulted in unequal utilization of existing schools and some unreasonably long travel distances by students. Good management of the rural schools is fundamental to meeting Chile's goal of providing quality education to its citizens. Seeking to improve the situation, the Ministry of Education ordered a study of the optimal location and size of rural schools with the general goals of reducing the number of lesser quality multigrade schools and reducing student travel distances while maintaining reasonable costs. This paper presents results of this study obtained using an integer linear program that has been embedded in a geographical information system. We present computational results for the entire country. Recommendations include where to open new rural schools as well as where to expand, reduce, close, or leave unchanged existing schools. We show how recommendations are sensitive to key parameters such as the cost of transportation.     

The design and evaluation of a tool to support software designers at the whiteboard

Despite the availability of advanced software design tools, software developers are known to frequently turn to the whiteboard when faced with a specific design problem. Free from any restrictions, they engage in an informal form of software design that relies heavily on sketching. However, whereas whiteboards afford flexibility and fluidity, they at the same time limit a designer in only being able to draw and erase content. This paper presents Calico, a novel software design tool that leverages electronic whiteboards to enhance the design experience with explicit support for the creative, exploratory aspects of design. Specifically, Calico offers a grid to partition the drawing space, scraps to organize and manipulate the content, and a reusable palette together with gesture-based input to address several natural behaviors exhibited by software designers, including frequent shifts in focus, use of low-detail models, use of a variety of notations, and a tendency to refine representations. To evaluate Calico, we performed a laboratory experiment involving the design of an educational traffic simulator. Through a three-pronged analysis examining the use of Calico’s features, design behaviors exhibited by the participants, and the nature of the design conversations, we show how Calico benefits the design process as it takes place at the whiteboard.