高维空间L1范数约束的一类数据稀疏距离测度学习算法与应用

现有一类分类算法通常采用经典欧氏测度描述样本间相似关系,然而欧氏测度不能较好地反映一些数据集样本的内在分布结构,从而影响这些方法对数据的描述能力.提出一种用于改善一类分类器描述性能的高维空间一类数据距离测度学习算法,与已有距离测度学习算法相比,该算法只需提供目标类数据,通过引入样本先验分布正则化项和L1范数惩罚的距离测度稀疏性约束,能有效解决高维空间小样本情况下的一类数据距离测度学习问题,并通过采用分块协调下降算法高效的解决距离测度学习的优化问题.学习的距离测度能容易的嵌入到一类分类器中,仿真实验结果表明采用学习的距离测度能有效改善一类分类器的描述性能,特别能够改善SVDD的描述能力,从而使得一类分类器具有更强的推广能力.     

A robust tracking method with adaptive local spatial sparse representation

In this paper, a robust visual tracking method is proposed based on local spatial sparse representation. In the proposed approach, the learned target template is sparsely and compactly expressed by forming local spatial and trivial samples dynamically. An adaptive multiple subspaces appearance model is developed to describe the target appearance and construct the candidate target templates during the tracking process. An effective selection strategy is then employed to select the optimal sparse solution and locate the target accurately in the next frame. The experimental results have demonstrated that our method can perform well in the complex and noisy visual environment, such as heavy occlusions, dramatic illumination changes, and large pose variations in the video. Copyright © 2015 John Wiley & Sons, Ltd.     

Numerical simulation of the performance of a snow fence with airfoil snow plates by FVM

The aim of this work is to analyze the efficiency of a snow fence with airfoil snow plates to avoid the snowdrift formation, to improve visibility and to prevent blowing snow disasters on highways and railways. In order to attain this objective, it is necessary to solve particle transport equations along with the turbulent fluid flow equations since there are two phases: solid phase (snow particles) and fluid phase (air). In the first place, the turbulent flow is modelled by solving the Reynolds-averaged Navier-Stokes (RANS) equations for incompressible viscous flows through the finite volume method (FVM) and then, once the flow velocity field has been determined, representative particles are tracked using the Lagrangian approach. Within the particle transport models, we have used a particle transport model termed as Lagrangian particle tracking model, where particulates are tracked through the flow in a Lagrangian way. The full particulate phase is modelled by just a sample of about 15,000 individual particles. The tracking is carried out by forming a set of ordinary differential equations in time for each particle, consisting of equations for position and velocity. These equations are then integrated using a simple integration method to calculate the behaviour of the particles as they traverse the flow domain. Finally, the conclusions of this work are exposed.     

Steady state numerical simulation of the particle collection efficiency of a new urban sustainable gravity settler using design of experiments by FVM

The aim of this work is to analyze the efficiency of a new sustainable urban gravity settler to avoid the solid particle transport, to improve the water waste quality and to prevent pollution problems due to rain water harvesting in areas with no drainage pavement. In order to get this objective, it is necessary to solve particle transport equations along with the turbulent fluid flow equations since there are two phases: solid phase (sand particles) and fluid phase (water). In the first place, the turbulent flow is modelled by solving the Reynolds-averaged Navier-Stokes (RANS) equations for incompressible viscous flows through the finite volume method (FVM) and then, once the flow velocity field has been determined, representative particles are tracked using the Lagrangian approach. Within the particle transport models, a particle transport model termed as Lagrangian particle tracking model is used, where particulates are tracked through the flow in a Lagrangian way. The full particulate phase is modelled by just a sample of about 2,000 individual particles. The tracking is carried out by forming a set of ordinary differential equations in time for each particle, consisting of equations for position and velocity. These equations are then integrated using a simple integration method to calculate the behaviour of the particles as they traverse the flow domain. The entire FVM model is built and the design of experiments (DOE) method was used to limit the number of simulations required, saving on the computational time significantly needed to arrive at the optimum configuration of the settler. Finally, conclusions of this work are exposed.     

基于BalanceCascade-GBDT算法的类别不平衡虚假评论识别方法

虚假评论是电商发展过程中一个无法避免的难题. 针对在线评论数据中样本类别不平衡情况,提出基于BalanceCascade-GBDT算法的虚假评论识别方法. BalanceCascade算法通过设置分类器的误报率逐步缩小大类样本空间,然后集成所有基分类器构建最终分类器. GBDT以其高准确性和可解释性被广泛应用于分类问题中,并且作为样本扰动不稳定算法,是十分合适的基分类模型. 模型基于Yelp评论数据集,采用AUC值作为评价指标,并与逻辑回归、随机森林以及神经网络算法进行对比,实验证明了该方法的有效性.     

A multiple template approach for robust tracking of fast motion target

Target tracking is very important in computer vision and related areas. It is usually difficult to accurately track fast motion target with appearance variations. Sometimes the tracking algorithms fail for heavy appearance variations. A multiple template method to track fast motion target with appearance changes is presented under the framework of appearance model with Kalman filter. Firstly, we construct a multiple template appearance model, which includes both the original template and templates affinely transformed from original one. Generally speaking, appearance variations of fast motion target can be covered by affine transformation. Therefore, the affine transform-enhanced templates match the target of appearance variations better than conventional models. Secondly, we present an improved Kalman filter for approximate estimating the motion trail of the target and a modified similarity evaluation function for exact matching. The estimation approach can reduce time complexity of the algorithm and keep accuracy in the meantime. Thirdly, we propose an adaptive scheme for updating template set to alleviate the drift problem. The scheme considers the following differences: the weight differences in two successive frames; different types of affine transformation applied to templates. Finally, experiments demonstrate that the proposed algorithm is robust to appearance variation of fast motion target and achieves real-time performance on middle/low-range computing platform.     

基于改进Adaboost的信用评价方法

网络借贷环境下基于Adaboost的信用评价方法具有较高的基分类器分歧度和样本误分代价。现有研究没有考虑分歧度和误分代价对基分类器样本权重的影响,从而降低了网络借贷信用评价结果的有效性。为此,提出一种基于改进Adaboost的信用评价方法。该方法根据基分类器的误分率,样本在不同基分类器上分类结果的分歧程度,以及样本的误分代价等因素,调整Adaboost模型的样本赋权策略,使得改进后的Adaboost模型能够对分类困难样本和误分代价高的样本实施有针对性的学习,从而提高网络借贷信用评价结果的有效性。基于拍拍贷平台数据的实验结果表明,提出的方法在分类精度和误分代价等方面显著优于传统的基于Adaboost的信用评价方法。     

The Sample Average Approximation Method Applied to Stochastic Routing Problems: A Computational Study

The sample average approximation (SAA) method is an approach for solving stochastic optimization problems by using Monte Carlo simulation. In this technique the expected objective function of the stochastic problem is approximated by a sample average estimate derived from a random sample. The resulting sample average approximating problem is then solved by deterministic optimization techniques. The process is repeated with different samples to obtain candidate solutions along with statistical estimates of their optimality gaps.We present a detailed computational study of the application of the SAA method to solve three classes of stochastic routing problems. These stochastic problems involve an extremely large number of scenarios and first-stage integer variables. For each of the three problem classes, we use decomposition and branch-and-cut to solve the approximating problem within the SAA scheme. Our computational results indicate that the proposed method is successful in solving problems with up to 2¹⁶⁹⁴ scenarios to within an estimated 1.0% of optimality. Furthermore, a surprising observation is that the number of optimality cuts required to solve the approximating problem to optimality does not significantly increase with the size of the sample. Therefore, the observed computation times needed to find optimal solutions to the approximating problems grow only linearly with the sample size. As a result, we are able to find provably near-optimal solutions to these difficult stochastic programs using only a moderate amount of computation time.     

Coarse-scale particle tracking approaches for contaminant transport in fractured rock

Random-walk particle tracking methods are frequently used for modeling contaminant transport, as relevant to radionuclide transport in fractured rock. Standard particle-tracking methods need to be modified for handling discontinuities in velocity and diffusion coefficients as at fracture–matrix interfaces, and handling these discontinuities accurately requires time steps much smaller than the diffusion time scale across narrow fracture apertures. In this work we present coarse-scale particle tracking methods that exploit the contrast in diffusivities between fracture and rock matrix to allow the use of time steps much larger than the diffusion time scale across fracture apertures. Thus, they reduce computational effort by several orders of magnitude. We develop two coarse-scale versions of the standard particle tracking method, one applicable to particles starting in the fracture, and another to particles starting in the rock matrix. The two methods can be used in combination to track particles through individual fractures, including the influence of matrix diffusion. The main advantage of our methods result from the computationally efficient treatment of (two-way) fracture–matrix particle transfer. These methods can also be combined with existing particle tracking approaches for complex advection–diffusion–dispersion in fractures to handle fracture–matrix interactions efficiently.     

Sub-sample swapping for sequential Monte Carlo approximation of high-dimensional densities in the context of complex object tracking

In this paper, we address the problem of complex object tracking using the particle filter framework, which essentially amounts to estimate high-dimensional distributions by a sequential Monte Carlo algorithm. For this purpose, we first exploit Dynamic Bayesian Networks to determine conditionally independent subspaces of the object’s state space, which allows us to independently perform the particle filter’s propagations and corrections over small spaces. Second, we propose a swapping process to transform the weighted particle set provided by the update step of the particle filter into a “new particle set” better focusing on high peaks of the posterior distribution. This new methodology, called Swapping-Based Partitioned Sampling, is proved to be mathematically sound and is successfully tested and validated on synthetic video sequences for single or multiple articulated object tracking.     

支持向量机及其在提高采收率潜力预测中的应用

提高采收率潜力分析的基础是进行提高采收率方法的潜力预测 .建立提高采收率潜力预测模型从统计学习的角度来看 ,实质是属于函数逼近问题 .本文首次将统计学习理论及支持向量机方法引入提高采收率方法的潜力预测中 .根据 Vapnik结构风险最小化原则 ,应尽量提高学习机的泛化能力 ,即由有效的训练集样本得到的小的误差能够保证对独立的测试集仍保持小的误差 .在本文所用较少样本条件下 ,支持向量机方法能够兼顾模型的通用性和推广性 ,具有较好的应用前景 .研究中采用的是综合正交设计法、油藏数值模拟和经济评价等方法生成的理论样本集     

Decision functions for chain classifiers based on Bayesian networks for multi-label classification

Multi-label classification problems require each instance to be assigned a subset of a defined set of labels. This problem is equivalent to finding a multi-valued decision function that predicts a vector of binary classes. In this paper we study the decision boundaries of two widely used approaches for building multi-label classifiers, when Bayesian network-augmented naive Bayes classifiers are used as base models: Binary relevance method and chain classifiers. In particular extending previous single-label results to multi-label chain classifiers, we find polynomial expressions for the multi-valued decision functions associated with these methods. We prove upper boundings on the expressive power of both methods and we prove that chain classifiers provide a more expressive model than the binary relevance method.     

Formalizing and solving the problem of clustering in MCDA

The topic of clustering has been widely studied in the field of Data Analysis, where it is defined as an unsupervised process of grouping objects together based on notions of similarity. Clustering in the field of Multi-Criteria Decision Aid (MCDA) has seen a few adaptations of methods from Data Analysis, most of them however using concepts native to that field, such as the notions of similarity and distance measures. As in MCDA we model the preferences of a decision maker over a set of decision alternatives, we can find more diverse ways of comparing them than in Data Analysis. As a result, these alternatives may also be arranged into different potential structures. In this paper we wish to formally define the problem of clustering in MCDA using notions that are native to this field alone, and highlight the different structures which we may try to uncover through this process. Following this we propose a method for finding these structures. As in any clustering problem, finding the optimal result in an exact manner is impractical, and so we propose a stochastic heuristic approach, which we validate through tests on a large set of artificially generated benchmarks.     

On stability of sampling-reconstruction models

A useful sampling-reconstruction model should be stable with respect to different kind of small perturbations, regardless whether they result from jitter, measurement errors, or simply from a small change in the model assumptions. In this paper we prove this result for a large class of sampling models. We define different classes of perturbations and present a way of quantifying the robustness of a model with respect to them. We also use the theory of localized frames to study the dual frame method for recovering the original signal from its samples.     

Robust inference strategy in the presence of measurement error

In this paper, we consider a statistical model where samples are subject to measurement errors. Further, we propose a shrinkage estimation strategy by using the maximum empirical likelihood estimator (MELE) as the base estimator. Our asymptotic results clearly demonstrate the superiority of our proposed shrinkage strategy over the MELE. Monte Carlo simulation results show that such a performance still holds in finite samples. We apply our method to real data set.     

A test problem for molecular dynamics integrators

We derive a test problem for evaluating the ability of time-steppingmethods to preserve the statistical properties of systems inmolecular dynamics. We consider a family of deterministic systemsconsisting of a finite number of particles interacting on acompact interval. The particles are given random initial conditionsand interact through instantaneous energy- and momentum-conservingcollisions. As the number of particles, the particle density,and the mean particle speed go to infinity, the trajectory ofa tracer particle is shown to converge to a stationary Gaussianstochastic process. We approximate this system by one describedby a system of ordinary differential equations and provide numericalevidence that it converges to the same stochastic process. Wesimulate the latter system with a variety of numerical integrators,including the symplectic Euler method, a fourth-order Runge-Kuttamethod, and an energyconserving step-and-project method. Weassess the methods' ability to recapture the system's limitingstatistics and observe that symplectic Euler performs significantlybetter than the others for comparable computational expense.     

Sample Average Approximation Method for Chance Constrained Programming: Theory and Applications

We study sample approximations of chance constrained problems. In particular, we consider the sample average approximation (SAA) approach and discuss the convergence properties of the resulting problem. We discuss how one can use the SAA method to obtain good candidate solutions for chance constrained problems. Numerical experiments are performed to correctly tune the parameters involved in the SAA. In addition, we present a method for constructing statistical lower bounds for the optimal value of the considered problem and discuss how one should tune the underlying parameters. We apply the SAA to two chance constrained problems. The first is a linear portfolio selection problem with returns following a multivariate lognormal distribution. The second is a joint chance constrained version of a simple blending problem. B.K. Pagnoncelli’s research was supported by CAPES and FUNENSEG. S. Ahmed’s research was partly supported by the NSF Award DMI-0133943. A. Shapiro’s research was partly supported by the NSF Award DMI-0619977.     

Optimal construction and rebalancing of index-tracking portfolios

Index funds aim to track the performance of a financial index, such as, e.g., the Standard?&?Poor’s?500 index. Index funds have become popular because they offer attractive risk-return profiles at low costs. The index-tracking problem considered in this paper consists of rebalancing the composition of the index fund’s tracking portfolio in response to new market information and cash deposits and withdrawals from investors such that the index fund’s tracking accuracy is maximized. In a frictionless market, maximum tracking accuracy is achieved by investing the index fund’s entire capital in a tracking portfolio that has the same normalized value development as the index. In the presence of transaction costs, which reduce the fund’s capital, one has to manage the trade-off between transaction costs and similarity in terms of normalized value developments. Existing mathematical programing formulations for the index-tracking problem do not optimize this trade-off explicitly, which may result in substantial transaction costs or tracking portfolios that differ considerably from the index in terms of normalized value development. In this paper, we present a mixed-integer linear programing formulation with a novel optimization criterion that directly considers the trade-off between transaction costs and similarity in terms of normalized value development. In an experiment based on a set of real-world problem instances, the proposed formulation achieves a considerably higher tracking accuracy than state-of-the-art formulations.     

Adaptively refined dynamic program for linear spline regression

The linear spline regression problem is to determine a piecewise linear function for estimating a set of given points while minimizing a given measure of misfit or error. This is a classical problem in computational statistics and operations research; dynamic programming was proposed as a solution technique more than 40 years ago by Bellman and Roth (J Am Stat Assoc 64:1079–1084, 1969). The algorithm requires a discretization of the solution space to define a grid of candidate breakpoints. This paper proposes an adaptive refinement scheme for the grid of candidate breakpoints in order to allow the dynamic programming method to scale for larger instances of the problem. We evaluate the quality of solutions found on small instances compared with optimal solutions determined by a novel integer programming formulation of the problem. We also consider a generalization of the linear spline regression problem to fit multiple curves that share breakpoint horizontal coordinates, and we extend our method to solve the generalized problem. Computational experiments verify that our nonuniform grid construction schemes are useful for computing high-quality solutions for both the single-curve and two-curve linear spline regression problem.     

Interval estimation of the mean of a normal distribution based on quantized observations

We consider the problem of making statistical inference about the mean of a normal distribution based on a random sample of quantized (digitized) observations. This problem arises, for example, in a measurement process with errors drawn from a normal distribution and with a measurement device or process with a known resolution, such as the resolution of an analog-to-digital converter or another digital instrument. In this paper we investigate the effect of quantization on subsequent statistical inference about the true mean. If the standard deviation of the measurement error is large with respect to the resolution of the indicating measurement device, the effect of quantization (digitization) diminishes and standard statistical inference is still valid. Hence, in this paper we consider situations where the standard deviation of the measurement error is relatively small. By Monte Carlo simulations we compare small sample properties of the interval estimators of the mean based on standard approach (i.e. by ignoring the fact that the measurements have been quantized) with some recently suggested methods, including the interval estimators based on maximum likelihood approach and the fiducial approach. The paper extends the original study by Hannig et al. (2007).