DiffPRFs:random forest under differential privacy

A differential privacy algorithm DiffPRFs based on random forests was proposed.Exponential mechanism was used to select split point and split attribute in each decision tree building process,and noise was added according to Laplace mechanism.Differential privacy protection requirement was satisfied through overall process.Compared to existed algorithms,the proposed method does not require pre-discretization of continuous attributes which significantly reduces the performance cost of preprocessing in large multi-dimensional dataset.Classification is achieved conveniently and efficiently while maintains the high accuracy.Experimental results demonstrate the effectiveness and superiority of the algorithm compared to other classification algorithms.     

Delay-dependent fault detection for switched linear systems with time-varying delays—the average dwell time approach

In this paper, the fault detection problem is investigated for a class of discrete-time switched linear systems with time-varying delays. The main purpose is to design a fault detection filter such that, for all unknown inputs, control inputs and time delays, the estimation error between the residual and fault is minimized in an exponential way. The fault detection problem is converted into an exponential H_∞ filtering problem. By using a newly constructed Lyapunov functional and the average dwell time scheme, a novel delay-dependent sufficient condition for the solvability of this problem is established in terms of linear matrix inequalities (LMIs). A numerical example is given to demonstrate the effectiveness of the developed theoretical results.     

Three-dimensional optical data storage through multi-photon confocal microscopy and imaging

Three dimensional optical data storage is one of the most promising tools to respond to the always growing demand for high data storage capacity. Here, we focused a femtosecond laser source by means of a confocal microscope onto different transparent recording media. The purpose of the study is to probe the capability of the system to independently address different data layers within the storage medium achieving thus three dimensional data storage. We demonstrated the possibility to write superposed independent layers of data due to either multiphoton excitation or to local optical breakdown and the performances observed in the different types of media used are compared.     

Entropy in Subordination and Filtering

We propose a stochastic model of transmitting random information at random time. In this model, the signal is observed as a random sampling according to an increasing stable stochastic process. Thus we are given a subordinate stochastic process which is a typical irreversible process. As the characteristic of this phenomea we observe the loss of entropy.     

Detecting and imaging dielectric objects from real data: A shape-based approach

In this paper we investigate the performance of a shape-reconstruction technique as tested on the ‘Marseille data’. This approach, which is based on a level set technique, offers several advantages compared to other approaches, as for example well-defined boundaries and the incorporation of an intrinsic regularization in the form of a priori assumptions regarding the general structures in the medium. The level set strategy (which is an implicit representation of the shapes) frees us from topological restrictions during this reconstruction process. Our algorithm is aiming at, not only detecting the objects, but simultaneously determining their approximate locations, sizes and dielectric properties. The numerical experiments show the utility of this method.     

Full‐duplex analogue/digital data transmission using lasers coupled to GI fibre and its application

This paper describes a low‐cost version of a full‐duplex optical fibre analogue/digital data transmission link whose practical implementation is simplified by using a pair of laser diodes, each having a built‐in photodiode for monitoring the laser output power, and its application to vibration test data transmission. The transmission link sends digital data of up to 9600 bps in one direction, and analogue data of DC to 100 kHz and image data specified by an NTSC colour video signal in the opposite direction, all at the same time through one GI optical fibre. Being simple in configuration and stable against noise, this type of link has a variety of applications within factory, home and offices. Copyright © 2001 John Wiley & Sons, Ltd.     

Time and direction of arrival detection and filtering for imaging in strongly scattering random media

We study detection and imaging of small reflectors in heavy clutter, using an array of transducers that emits and receives sound waves. Heavy clutter means that multiple scattering of the waves in the heterogeneous host medium is strong and overwhelms the arrivals from the small reflectors. Building on the adaptive time-frequency filter of [Borcea et al, SIIMS 2011;4(3)], we propose a robust method for detecting the direction of arrival of the direct echoes from the small reflectors, and suppressing the unwanted clutter backscatter. This improves the resolution of imaging. We illustrate the performance of the method with realistic numerical simulations in a non-destructive testing setup.     

Analyzing the Sensitivity of Generalized Linear Models to Incomplete Outcomes via the IDE Algorithm

Incomplete data models typically involve strong untestable assumptions about the missing data distribution. As inference may critically depend on them, the importance of sensitivity analysis is well recognized. Molenberghs, Kenward, and Goetghebeur proposed a formal frequentist approach to sensitivity analysis which distinguishes ignorance due to unintended incompleteness from imprecision due to finite sampling by design. They combine both sources of variation into uncertainty. This article develops estimation tools for ignorance and uncertainty concerning regression coefficients in a complete data model when some of the intended outcome values are missing. Exhaustive enumeration of all possible imputations for the missing data requires enormous computational resources. In contrast, when the boundary of the occupied region is of greatest interest, reasonable computational efficiency may be achieved via the imputation towards directional extremes (IDE) algorithm. This is a special imputation method designed to mark the boundary of the region by maximizing the direction of change of the complete data estimator caused by perturbations to the imputed outcomes. For multi-dimensional parameters, a dimension reduction approach is considered. Additional insights are obtained by considering structures within the region, and by introducing external knowledge to narrow the boundary to useful proportions. Special properties hold for the generalized linear model. Examples from a Kenyan HIV study will illustrate the points.     

基于加权相似性的BIRCH聚类算法

BIRCH方法是一个集成的层次聚类方法.它克服了凝聚层次聚类方法所面临的两个难点:可伸缩性和不能撤销前一步工作的问题.基于BIRCH聚类的多阶段聚类算法思想,结合基于权重的欧式距离度量和基于划分的K-means算法,提出了一种基于加权相似性的BIRCH聚类方法,并将方法应用在时间序列的气象数据分析中.