Searcharxiv⌕ Search

arXiv subjects

Yong Xia

Publications and source records attributed to Yong Xia.

At least 163 records · Page 9Linked to original sources
arXiv2018

A Pulmonary Nodule Detection Model Based on Progressive Resolution and Hierarchical Saliency

Detection of pulmonary nodules on chest CT is an essential step in the early diagnosis of lung cancer, which is critical for best patient care. Although a number of computer-aided nodule detection methods have been published in the literature, these methods still have two major drawbacks: missing out true nodules during the detection of nodule candidates and less-accurate identification of nodules from non-nodule. In this paper, we propose an automated pulmonary nodule detection algorithm that jointly combines progressive resolution and hierarchical saliency. Specifically, we design a 3D progressive resolution-based densely dilated FCN, namely the progressive resolution network (PRN), to detect nodule candidates inside the lung, and construct a densely dilated 3D CNN with hierarchical saliency, namely the hierarchical saliency network (HSN), to simultaneously identify genuine nodules from those candidates and estimate the diameters of nodules. We evaluated our algorithm on the benchmark LUng Nodule Analysis 2016 (LUNA16) dataset and achieved a state-of-the-art detection score. Our results suggest that the proposed algorithm can effectively detect pulmonary nodules on chest CT and accurately estimate their diameters.

cs.CV↗
arXiv2018

Autonomous Deep Learning: A Genetic DCNN Designer for Image Classification

Recent years have witnessed the breakthrough success of deep convolutional neural networks (DCNNs) in image classification and other vision applications. Although freeing users from the troublesome handcrafted feature extraction by providing a uniform feature extraction-classification framework, DCNNs still require a handcrafted design of their architectures. In this paper, we propose the genetic DCNN designer, an autonomous learning algorithm can generate a DCNN architecture automatically based on the data available for a specific image classification problem. We first partition a DCNN into multiple stacked meta convolutional blocks and fully connected blocks, each containing the operations of convolution, pooling, fully connection, batch normalization, activation and drop out, and thus convert the architecture into an integer vector. Then, we use refined evolutionary operations, including selection, mutation and crossover to evolve a population of DCNN architectures. Our results on the MNIST, Fashion-MNIST, EMNISTDigit, EMNIST-Letter, CIFAR10 and CIFAR100 datasets suggest that the proposed genetic DCNN designer is able to produce automatically DCNN architectures, whose performance is comparable to, if not better than, that of stateof- the-art DCNN models

cs.CV↗
arXiv2018

Optical stimulated slowing of polar heavy-atom molecules with a constant beat phase

Polar heavy-atom molecules have been well recognized as promising candidates for precision measurements and tests of fundamental physics. A much slower molecular beam to increase the interaction time should lead to a more sensitive measurement. Here we theoretically demonstrate the possibility of the stimulated longitudinal slowing of heavy-atom molecules by the coherent optical bichromatic force with a constant beat phase. Taking the YbF meolecule as an example, we show that a rapid and short-distance deceleration of heavy molecules by a phase-compensation method is feasible with moderate conditions. A molecular beam of YbF with a forward velocity of 120 m/s can be decelerated below 10 m/s within a distance of 3.5 cm and with a laser irradiance for each traveling wave of 107.2 W/cm$^2$. We also give a simple approach to estimate the performance of the BCF on some other heavy molecules, which is helpful for making a rapid evaluation on the feasibility of the stimulated slowing experiment. Our proposed slowing method could be a promising approach to break through the space constraint or the limited capture efficiency of molecules loadable into a MOT in traditional deceleration schemes, opening the possibility for a significant improvement of the precision measurement sensitivity.

physics.atom-ph↗
arXiv2018

An efficient global optimization algorithm for maximizing the sum of two generalized Rayleigh quotients

Maximizing the sum of two generalized Rayleigh quotients (SRQ) can be reformulated as a one-dimensional optimization problem, where the function value evaluations are reduced to solving semi-definite programming (SDP) subproblems. In this paper, we first use the dual SDP subproblem to construct an explicit overestimation and then propose a branch-and-bound algorithm to globally solve (SRQ). Numerical results demonstrate that it is even more efficient than the recent SDP-based heuristic algorithm.

math.OC↗
arXiv2017

Classification of Medical Images and Illustrations in the Biomedical Literature Using Synergic Deep Learning

The Classification of medical images and illustrations in the literature aims to label a medical image according to the modality it was produced or label an illustration according to its production attributes. It is an essential and challenging research hotspot in the area of automated literature review, retrieval and mining. The significant intra-class variation and inter-class similarity caused by the diverse imaging modalities and various illustration types brings a great deal of difficulties to the problem. In this paper, we propose a synergic deep learning (SDL) model to address this issue. Specifically, a dual deep convolutional neural network with a synergic signal system is designed to mutually learn image representation. The synergic signal is used to verify whether the input image pair belongs to the same category and to give the corrective feedback if a synergic error exists. Our SDL model can be trained 'end to end'. In the test phase, the class label of an input can be predicted by averaging the likelihood probabilities obtained by two convolutional neural network components. Experimental results on the ImageCLEF2016 Subfigure Classification Challenge suggest that our proposed SDL model achieves the state-of-the art performance in this medical image classification problem and its accuracy is higher than that of the first place solution on the Challenge leader board so far.

cs.CV↗
arXiv2017

A Tribe Competition-Based Genetic Algorithm for Feature Selection in Pattern Classification

Feature selection has always been a critical step in pattern recognition, in which evolutionary algorithms, such as the genetic algorithm (GA), are most commonly used. However, the individual encoding scheme used in various GAs would either pose a bias on the solution or require a pre-specified number of features, and hence may lead to less accurate results. In this paper, a tribe competition-based genetic algorithm (TCbGA) is proposed for feature selection in pattern classification. The population of individuals is divided into multiple tribes, and the initialization and evolutionary operations are modified to ensure that the number of selected features in each tribe follows a Gaussian distribution. Thus each tribe focuses on exploring a specific part of the solution space. Meanwhile, tribe competition is introduced to the evolution process, which allows the winning tribes, which produce better individuals, to enlarge their sizes, i.e. having more individuals to search their parts of the solution space. This algorithm, therefore, avoids the bias on solutions and requirement of a pre-specified number of features. We have evaluated our algorithm against several state-of-the-art feature selection approaches on 20 benchmark datasets. Our results suggest that the proposed TCbGA algorithm can identify the optimal feature subset more effectively and produce more accurate pattern classification.

cs.LG↗
arXiv2016

Approximation of the weighted maximin dispersion problem over Lp-ball: SDP relaxation is misleading

Consider the problem of finding a point in a unit $n$-dimensional $\ell_p$-ball ($p\ge 2$) such that the minimum of the weighted Euclidean distance from given $m$ points is maximized. We show in this paper that the recent SDP-relaxation-based approximation algorithm [SIAM J. Optim. 23(4), 2264-2294, 2013] will not only provide the first theoretical approximation bound of $\frac{1-O\left(\sqrt{ \ln(m)/n}\right)}{2}$, but also perform much better in practice, if the SDP relaxation is removed and the optimal solution of the SDP relaxation is replaced by a simple scalar matrix.

math.OC↗
arXiv2016

On the Ball-Constrained Weighted Maximin Dispersion Problem

The ball-constrained weighted maximin dispersion problem $(\rm P_{ball})$ is to find a point in an $n$-dimensional Euclidean ball such that the minimum of the weighted Euclidean distance from given $m$ points is maximized. We propose a new second-order cone programming relaxation for $(\rm P_{ball})$. Under the condition $m\le n$, $(\rm P_{ball})$ is polynomial-time solvable since the new relaxation is shown to be tight. In general, we prove that $({\rm P_{ball}})$ is NP-hard. Then, we propose a new randomized approximation algorithm for solving $({\rm P_{ball}})$, which provides a new approximation bound of $\frac{1-O(\sqrt{\ln(m)/n})}{2}$.

math.OC↗
arXiv2015

Similar Handwritten Chinese Character Discrimination by Weakly Supervised Learning

Traditional approaches for handwritten Chinese character recognition suffer in classifying similar characters. In this paper, we propose to discriminate similar handwritten Chinese characters by using weakly supervised learning. Our approach learns a discriminative SVM for each similar pair which simultaneously localizes the discriminative region of similar character and makes the classification. For the first time, similar handwritten Chinese character recognition (SHCCR) is formulated as an optimization problem extended from SVM. We also propose a novel feature descriptor, Gradient Context, and apply bag-of-words model to represent regions with different scales. In our method, we do not need to select a sized-fixed sub-window to differentiate similar characters. The unconstrained property makes our method well adapted to high variance in the size and position of discriminative regions in similar handwritten Chinese characters. We evaluate our proposed approach over the CASIA Chinese character data set and the results show that our method outperforms the state of the art.

cs.CV↗
arXiv2015

Uniform Quadratic Optimization and Extensions

The uniform quadratic optimizatin problem (UQ) is a nonconvex quadratic constrained quadratic programming (QCQP) sharing the same Hessian matrix. Based on the second-order cone programming (SOCP) relaxation, we establish a new sufficient condition to guarantee strong duality for (UQ) and then extend it to (QCQP), which not only covers several well-known results in literature but also partially gives answers to a few open questions. For convex constrained nonconvex (UQ), we propose an improved approximation algorithm based on (SOCP). Our approximation bound is dimensional independent. As an application, we establish the first approximation bound for the problem of finding the Chebyshev center of the intersection of several balls.

math.OC↗
arXiv2015

S-Lemma with Equality and Its Applications

Let $f(x)=x^TAx+2a^Tx+c$ and $h(x)=x^TBx+2b^Tx+d$ be two quadratic functions having symmetric matrices $A$ and $B$. The S-lemma with equality asks when the unsolvability of the system $f(x)<0, h(x)=0$ implies the existence of a real number $μ$ such that $f(x) + μh(x)\ge0, ~\forall x\in \mathbb{R}^n$. The problem is much harder than the inequality version which asserts that, under Slater condition, $f(x)<0, h(x)\le0$ is unsolvable if and only if $f(x) + μh(x)\ge0, ~\forall x\in \mathbb{R}^n$ for some $μ\ge0$. In this paper, we show that the S-lemma with equality does not hold only when the matrix $A$ has exactly one negative eigenvalue and $h(x)$ is a non-constant linear function ($B=0, b\not=0$). As an application, we can globally solve $\inf\{f(x)\vert h(x)=0\}$ as well as the two-sided generalized trust region subproblem $\inf\{f(x)\vert l\le h(x)\le u\}$ without any condition. Moreover, the convexity of the joint numerical range $\{(f(x), h_1(x),\ldots, h_p(x)):~x\in\Bbb R^n\}$ where $f$ is a (possibly non-convex) quadratic function and $h_1(x),\ldots,h_p(x)$ are affine functions can be characterized using the newly developed S-lemma with equality.

math.OC↗
arXiv2014

An Improved Analysis of Semidefinite Approximation Bound for Nonconvex Nonhomogeneous Quadratic Optimization with Ellipsoid Constraints

We consider the problem of approximating nonconvex quadratic optimization with ellipsoid constraints (ECQP). We show some SDP-based approximation bounds for special cases of (ECQP) can be improved by trivially applying the extened Pataki's procedure. The main result of this paper is to give a new analysis on approximating (ECQP) by the SDP relaxation, which greatly improves Tseng's result [SIAM Journal Optimization, 14, 268-283, 2003]. As an application, we strictly improve the approximation ratio for the assignment-polytope constrained quadratic program.

math.OC↗
arXiv2014

Double Well Potential Function and Its Optimization in the n-dimensional Real Space - Part II

In contrast to taking the dual approach for finding a global minimum solution of a double well potential function, in Part II of the paper, we characterize a local minimizer, local maximizer, and global minimizer directly from the primal side. It is proven that, for a ``nonsingular" double well function, there exists at most one local, but non-global, minimizer and at most one local maximizer. Moreover, when it exists, the local maximizer is ``surrounded" by local minimizers in the sense that the norm of the local maximizer is strictly less than that of any local minimizer. We also establish some necessary and sufficient optimality conditions for the global minimizer, local non-global minimizer and local maximizer by studying a convex secular function over specific intervals. These conditions lead to three algorithms for identifying different types of critical points of a given double well function.

math.OC↗
arXiv2014

An SDP Approach For Solving Quadratic Fractional Programming Problems

This paper considers a fractional programming problem (P) which minimizes a ratio of quadratic functions subject to a two-sided quadratic constraint. As is well-known, the fractional objective function can be replaced by a parametric family of quadratic functions, which makes (P) highly related to, but more difficult than a single quadratic programming problem subject to a similar constraint set. The task is to find the optimal parameter $λ^*$ and then look for the optimal solution if $λ^*$ is attained. Contrasted with the classical Dinkelbach method that iterates over the parameter, we propose a suitable constraint qualification under which a new version of the S-lemma with an equality can be proved so as to compute $λ^*$ directly via an exact SDP relaxation. When the constraint set of (P) is degenerated to become an one-sided inequality, the same SDP approach can be applied to solve (P) {\it without any condition}. We observe that the difference between a two-sided problem and an one-sided problem lies in the fact that the S-lemma with an equality does not have a natural Slater point to hold, which makes the former essentially more difficult than the latter. This work does not, either, assume the existence of a positive-definite linear combination of the quadratic terms (also known as the dual Slater condition, or a positive-definite matrix pencil), our result thus provides a novel extension to the so-called "hard case" of the generalized trust region subproblem subject to the upper and the lower level set of a quadratic function.

math.OC↗
arXiv2013

A new semidefinite relaxation for $\ell_{1}$-constrained quadratic optimization and extensions

In this paper, by improving the variable-splitting approach, we propose a new semidefinite programming (SDP) relaxation for the nonconvex quadratic optimization problem over the $\ell_1$ unit ball (QPL1). It dominates the state-of-the-art SDP-based bound for (QPL1). As extensions, we apply the new approach to the relaxation problem of the sparse principal component analysis and the nonconvex quadratic optimization problem over the $\ell_p$ ($1< p<2$) unit ball and then show the dominance of the new relaxation.

math.OC↗