Resource Type

Journal Article 70

Year

2023 6

2022 8

2021 5

2020 7

2019 4

2018 7

2017 3

2016 3

2015 3

2014 2

2013 2

2009 1

2007 1

2006 2

2005 2

2004 4

2003 4

2002 4

2001 1

1999 1

open ︾

Keywords

Blockchain 3

coherent structures 2

motion compensation 2

AOD 1

Acoustic localization 1

Adaptive beamforming 1

Adaptive control 1

Adaptive observers 1

Aerosol pollution levels 1

Aerosol type 1

Air entrainment 1

Aircraft testing 1

Amorphous solid 1

Anonymity 1

Atomic clusters 1

Automated scanning 1

Automatic modulation classification 1

Bandwith allocation 1

Bankcard enrollment 1

open ︾

Search scope:

排序： Display mode:

Block coordinate descent with time perturbation for nonconvex nonsmooth problems in real-world studies Research Article

Rui Liu, Wei-chu Sun, Tao Hou, Chun-hong Hu, Lin-bo Qiao,liuruirui@csu.edu.cn,smsysun@foxmail.com,houtao@csu.edu.cn,huchunhong@csu.edu.cn,qiao.linbo@nudt.edu.cn

Frontiers of Information Technology & Electronic Engineering 2019, Volume 20, Issue 10,   Pages 1390-1403 doi: 10.1631/FITEE.1900341

Abstract: The era of big data in healthcare is here, and this era will significantly improve medicine and especially oncology. However, traditional machine learning algorithms need to be promoted to solve such large-scale real-world problems due to a large amount of data that needs to be analyzed and the difficulty in solving problems with nonconvex nonlinear settings. We aim to minimize the composite of a smooth nonlinear function and a block-separable nonconvex function on a large number of block variables with inequality constraints. We propose a novel parallel first-order optimization method, called asynchronous block coordinate descent with (ATP), which adopts a technique that escapes from saddle points and sub-optimal local points. The details of the proposed method are presented with analyses of convergence and iteration complexity properties. Experiments conducted on real-world machine learning problems validate the efficacy of our proposed method. The experimental results demonstrate that enables ATP to escape from saddle points and sub-optimal points, providing a promising way to handle nonconvex optimization problems with inequality constraints employing asynchronous block coordinate descent. The asynchronous parallel implementation on shared memory multi-core platforms indicates that the proposed algorithm, ATP, has strong scalability.

Keywords： 收敛分析；异步块坐标下降法；时间扰动；非凸非平滑优化；真实世界研究    

HTML PDF Collect

HAM: a deep collaborative ranking method incorporating textual information Research Articles

Cheng-wei Wang, Teng-fei Zhou, Chen Chen, Tian-lei Hu, Gang Chen,rr@zju.edu.cn,zhoutengfei@zju.edu.cn,cc33@zju.edu.cn,htl@zju.edu.cn,cg@zju.edu.cn

Frontiers of Information Technology & Electronic Engineering 2020, Volume 21, Issue 8,   Pages 1119-1266 doi: 10.1631/FITEE.1900382

Abstract: The recommendation task with a textual corpus aims to model customer preferences from both user feedback and item textual descriptions. It is highly desirable to explore a very deep neural network to capture the complicated nonlinear preferences. However, training a deeper recommender is not as effortless as simply adding layers. A deeper recommender suffers from the gradient vanishing/exploding issue and cannot be easily trained by gradient-based methods. Moreover, textual descriptions probably contain noisy word sequences. Directly extracting feature vectors from them can harm the recommender’s performance. To overcome these difficulties, we propose a new recommendation method named the HighwAy recoMmender (HAM). HAM explores a highway mechanism to make gradient-based training methods stable. A multi-head attention mechanism is devised to automatically denoise textual information. Moreover, a method is devised to train a deep neural recommender. Empirical studies show that the proposed method outperforms state-of-the-art methods significantly in terms of accuracy.

Keywords： 深度学习；推荐系统；高速公路网络；块坐标梯度下降    

HTML PDF Collect

A descent method for the Dubins traveling salesman problem with neighborhoods Research Articles

Zheng Chen, Chen-hao Sun, Xue-ming Shao, Wen-jie Zhao,z_chen@zju.edu.cn,mecsxm@zju.edu.cn

Frontiers of Information Technology & Electronic Engineering 2021, Volume 22, Issue 5,   Pages 615-766 doi: 10.1631/FITEE.2000041

Abstract: In this study, we focus mainly on the problem of finding the minimum-length path through a set of circular regions by a fixed-wing unmanned aerial vehicle. Such a problem is referred to as the with neighborhoods (DTSPN). Algorithms developed in the literature for solving DTSPN either are computationally demanding or generate low-quality solutions. To achieve a better trade-off between solution quality and computational cost, an efficient gradient-free is designed. The core idea of the is to decompose DTSPN into a series of subproblems, each of which consists of finding the minimum-length path of a from a configuration to another configuration via an intermediate circular region. By analyzing the geometric properties of the subproblems, we use a bisection method to solve the subproblems. As a result, the can efficiently address DTSPN by successively solving a series of subproblems. Finally, several numerical experiments are carried out to demonstrate the in comparison with several existing algorithms.

Keywords： Dubins飞行器；坐标下降法；Dubins旅行商问题    

HTML PDF Collect

Improved Matching Algorithms for Linear Face Class Model

Fu Yun,Zheng Nanning

Strategic Study of CAE 2005, Volume 7, Issue 2,   Pages 47-56

Abstract:

An advanced matching technique for linear face class model is proposed, which can solve the problem of detailed controlling and robust iteration for the realistic facial modeling. A new method——Dynamic Gaussian Pyramid Analysis (DGPA), which combines Non-Uniform Sampling (NUS) method and Multi-Resolution Analysis, is presented. Integrating the PS Sampling and the Cluster Random Sampling, the distribution of the sampled points in each level images of the Gaussian pyramid is adjusted dynamically. In coarse-to-fine scheme, the minimization algorithm is used to compute the near global optimal solution that may fit to yield accurate model matching. Dynamic adjusting the boundary of the sampling cluster area and the resampling ratio, the detailed representations are effectively controlled, and the model creation is quite robust. An improved Stochastic Gradient Descent (SGD) algorithm based on the Correlative Disturbance (CD) and Adaptive Learning Rate (ALR) is exploited to accelerate iteration convergence and compute valid model parameters. With the examples of MPI Caucasian Face and AI&R Asian Face databases, experimental results in subjective evaluation and objective evaluation demonstrate the advanced model matching technique.

Keywords： facial modeling     model matching     stochastic gradient descent     non-uniform sampling     multiresolution analysis    

HTML PDF Collect

Application of Satellite Gravity Gradiometry Data to the Refinement of the Earth’s Gravity Field

Ning Jinsheng,Luo Zhicai,Chen Yongqi

Strategic Study of CAE 2002, Volume 4, Issue 7,   Pages 23-28

Abstract:

In recent times, one of the primary scientific objectives of physical geodesy is to determine the geoid with centimeter level and to develop the ultra high global earth’s gravity field model，and for this purpose satellite gravity gradiometry is one of the most promising techniques. This paper first comments on the main progress of satellite gravity gradiometry, and then discusses some theories and methods for refining the earth´s gravity field using satellite gravity gradiometry data.

Keywords： satellite gravity gradiometry     satellite gravity gradiometry boundary value problem     earth's gravity field model     geoid    

HTML PDF Collect

Study on Down Knock-in Renewal Option Pricing

Wu Yun,He Jianmin

Strategic Study of CAE 2004, Volume 6, Issue 12,   Pages 47-50

Abstract:

In this paper, firstly, the basic connotation of modern leasing and the basic principle of real option are dissertated. Then, the down knock-in renewal option pricing model and its solution are put forward. At last, example is provided.

Keywords： leasing     renewal     leasing option     real option     down knock-in option    

HTML PDF Collect

Numerical Computation of Flow and Pollutant Diffusion - Transportation With Non-orthogonal Curvilinear Coordinates

Wu Xiugang,Shen Yongming,Zheng Yonghong,Yang Zhifeng

Strategic Study of CAE 2003, Volume 5, Issue 2,   Pages 57-61

Abstract:

Based on depth-averaged mathematic model of flow and pollutant transportation in non-orthogonal curvilinear coordinates, the k-ε double equations´ turbulence model and the coupling arithmetic of velocity-water depth are adopted, which are applied to the numerical simulation of the flow and the contamination diffusion-transportation. The flow field and concentration of discharge both along bank and in the centerline of the meandering channel in laboratory are calculated, and the calculating results accord well with the measured data.

Keywords： curvilinear coordinates     meandering channel     diffusion-transportation     IMPLEC algorithm     concentration field    

HTML PDF Collect

The Ejection and Sweep in the Wall Region of a Turbulent Boundary Layer Under Adverse Pressure Gradient

Zhang Qiang,Lu Lipeng

Strategic Study of CAE 2003, Volume 5, Issue 11,   Pages 47-50

Abstract:

The evolution of coherent structures in the wall region of a turbulent boundary layer under adverse pressure gradients was studied by using DNS method. The motions of the second and the fourth quadrant were analyzed. The results show that to the shear stress, the contribution of sweep motions is stronger than that of ejections under adverse pressure gradient. And the sweeps contribute to the transport of shear-stress and kinetic energy of turbulence more strongly than the ejections under adverse pressure gradient.

Keywords： coherent structures     turbulent boundary layer     adverse pressure gradients     direct numerical simulation    

HTML PDF Collect

Fractional-order global optimal backpropagation machine trained by an improved fractional-order steepest descent method Research Articles

Yi-fei Pu, Jian Wang,puyifei@scu.edu.cn,wangjiannl@upc.edu.cn

Frontiers of Information Technology & Electronic Engineering 2020, Volume 21, Issue 6,   Pages 809-962 doi: 10.1631/FITEE.1900593

Abstract: We introduce the fractional-order global optimal backpropagation machine, which is trained by an improved (FSDM). This is a fractional-order backpropagation neural network (FBPNN), a state-of-the-art fractional-order branch of the family of backpropagation neural networks (BPNNs), different from the majority of the previous classic first-order BPNNs which are trained by the traditional first-order steepest descent method. The reverse incremental search of the proposed FBPNN is in the negative directions of the approximate fractional-order partial derivatives of the square error. First, the theoretical concept of an FBPNN trained by an improved FSDM is described mathematically. Then, the mathematical proof of fractional-order global optimal convergence, an assumption of the structure, and of the FBPNN are analyzed in detail. Finally, we perform three (types of) experiments to compare the performances of an FBPNN and a classic first-order BPNN, i.e., example function approximation, , and comparison of global search and error fitting abilities with real data. The higher optimal search ability of an FBPNN to determine the global optimal solution is the major advantage that makes the FBPNN superior to a classic first-order BPNN.

Keywords： 分数阶微积分；分数阶反向传播算法；分数阶最速下降法；均方误差；分数阶多尺度全局优化    

HTML PDF Collect

Reinforcement design of anchor blocks for external tendons

Hong Hao,Tong Yuqiang ,He Zhiqi,Liu Zhao

Strategic Study of CAE 2013, Volume 15, Issue 8,   Pages 95-98

Abstract:

Reinforced blocks are common anchorage forms for external tendons. However, there was no mature design method. Through finite element analysis, the paper revealed that the tensile stress concentration was caused by three typical local actions. On this basis, three equations for calculating resultant forces of these local actions were formulated. Thus a quantitative reinforcement design method of anchor blocks was proposed. Finally, an anchor block of the Fourth Nanjing Yangtze River Bridge was presented as a design example to demonstrate the effectiveness and convenience of the proposed method.

Keywords： anchor block     external prestressing     reinforcement design     bursting force     cantilever action     tie-back    

HTML PDF Collect

The Nanoscale Density Gradient as a Structural Stabilizer for Glass Formation Article

Shaoxiong Zhou, Bangshao Dong, Yanguo Wang, Jingyu Qin, Weihua Wang

Engineering 2023, Volume 29, Issue 10,   Pages 120-129 doi: 10.1016/j.eng.2023.01.010

Abstract:

The rapid cooling of a metallic liquid (ML) results in short-range order (SRO) among the atomic arrangements and a disordered structure in the resulting metallic glass (MG). These phenomena cause various possible features in the microscopic structure of the MG, presenting a puzzle about the nature of the MGs' microscopic structure beyond SRO. In this study, the nanoscale density gradient (NDG) originating from a sequential arrangement of clusters with different atomic packing densities (APDs), representing the medium-range structural heterogeneity in Zr60Cu30Al10 MG, was characterized using electron tomography (ET) combined with image simulations based on structure modeling. The coarse polyhedrons with distinct facets identified in the three-dimensional images coincide with icosahedron-like clusters and represent the spatial positions of clusters with high APDs. Rearrangements of the different clusters according to descending APD order in the glass-forming process are responsible for the NDG that stabilizes both the supercooled ML and the amorphous states and acts as a hidden rule in the transition from ML to MG.

Keywords： Rapid cooling     Amorphous solid     Density gradient     Electron tomography     Atomic clusters    

HTML PDF Collect

A Weighted Block-matching Criterion for the Hardware Implementation of Motion Estimators

Zhang Xia,Zheng Nanning,Zhang Guanglie,Wu Yong,Wang Shaorui,Xu Weipu

Strategic Study of CAE 2002, Volume 4, Issue 1,   Pages 47-53

Abstract:

In all kinds of digital video processing algorithms, motion compensated algorithm can acquire perfect performance because the motion information in the video signal has been considered. The hardware implementation of the motion estimator is the core of the various motion compensated digital video processings, which will be applied in real systems. Block-matching motion estimating algorithm is widely used in real system because of its low computing complication, easy realization and high call frequency of the block-matching criterion in the hardware systems. A new criterion called weighted minimized maximum error is proposed in this paper. This criterion can reduce the complexity of the motion estimator, decrease the area of the hardware and increase the speed of the hardware. On the other hand, the criterion is suitable to be applied to the recursive searching strategy which has an inherent weakness called error propagation.

Keywords： video processing     motion compensation     motion estimation     block-matching criterion    

HTML PDF Collect

Survey on Extraction Methods of Transition Region

Liu Suolan,Yang Jingyu

Strategic Study of CAE 2007, Volume 9, Issue 9,   Pages 89-96

Abstract:

Image segmentation is treated as a key issue in image processing and machine vision,  and has been the bottleneck of development.  Transition region is a special region located between the object and background.  Having the aid of extraction of transition region to segment image is a kind of burgeoning technology.  Two kinds of methods have been introduced: methods based on gradient and methods based on non-gradient.  Briefly,  the extracted effects and existent problem have been analyzed.

Keywords： transition region     extraction     image segmentation     gradient method     non-gradient method    

HTML PDF Collect

Quantifying multiple social relationships based on a multiplex stochastic block model Science Letter

Mincheng Wu, Zhen Li, Cunqi Shao, Shibo He,s18he@zju.edu.cn

Frontiers of Information Technology & Electronic Engineering 2021, Volume 22, Issue 11,   Pages 1458-1462 doi: 10.1631/FITEE.2000617

Abstract: Online s have attracted great attention recently, because they make it easy to build social connections for people all over the world. However, the observed structure of an online is always the aggregation of multiple social relationships. Thus, it is of great importance for real-world networks to reconstruct the full network structure using limited observations. The multiplex is introduced to describe multiple social ties, where different layers correspond to different attributes (e.g., age and gender of users in a ). In this letter, we aim to improve the model precision using maximum likelihood estimation, where the precision is defined by the cross entropy of parameters between the data and model. Within this framework, the layers and partitions of nodes in a are determined by natural node annotations, and the aggregate of the is available. Because the original has a high degree of freedom, we add an independent functional layer to cover it, and theoretically provide the optimal block number of the added layer. Empirical results verify the effectiveness of the proposed method using four measures, i.e., error of link probability, cross entropy, area under the receiver operating characteristic curve, and Bayes factor.

Keywords： 社交网络；多重网络；随机块模型    

HTML PDF Collect

A distributed stochastic optimization algorithm with gradient-tracking and distributed heavy-ball acceleration Research Articles

Bihao Sun, Jinhui Hu, Dawen Xia, Huaqing Li,huaqingli@swu.edu.cn

Frontiers of Information Technology & Electronic Engineering 2021, Volume 22, Issue 11,   Pages 1463-1476 doi: 10.1631/FITEE.2000615

Abstract: has been well developed in recent years due to its wide applications in machine learning and signal processing. In this paper, we focus on investigating to minimize a global objective. The objective is a sum of smooth and strongly convex local cost functions which are distributed over an undirected network of nodes. In contrast to existing works, we apply a distributed heavy-ball term to improve the convergence performance of the proposed algorithm. To accelerate the convergence of existing distributed stochastic first-order gradient methods, a momentum term is combined with a gradient-tracking technique. It is shown that the proposed algorithm has better acceleration ability than GT-SAGA without increasing the complexity. Extensive experiments on real-world datasets verify the effectiveness and correctness of the proposed algorithm.

Keywords： 分布式优化；高性能算法；多智能体系统；机器学习问题；随机梯度    

HTML PDF Collect

Title Author Date Type Operation

Block coordinate descent with time perturbation for nonconvex nonsmooth problems in real-world studies

Rui Liu, Wei-chu Sun, Tao Hou, Chun-hong Hu, Lin-bo Qiao,liuruirui@csu.edu.cn,smsysun@foxmail.com,houtao@csu.edu.cn,huchunhong@csu.edu.cn,qiao.linbo@nudt.edu.cn

Journal Article

HAM: a deep collaborative ranking method incorporating textual information

Cheng-wei Wang, Teng-fei Zhou, Chen Chen, Tian-lei Hu, Gang Chen,rr@zju.edu.cn,zhoutengfei@zju.edu.cn,cc33@zju.edu.cn,htl@zju.edu.cn,cg@zju.edu.cn

Journal Article

A descent method for the Dubins traveling salesman problem with neighborhoods

Zheng Chen, Chen-hao Sun, Xue-ming Shao, Wen-jie Zhao,z_chen@zju.edu.cn,mecsxm@zju.edu.cn

Journal Article

Improved Matching Algorithms for Linear Face Class Model

Fu Yun,Zheng Nanning

Journal Article

Application of Satellite Gravity Gradiometry Data to the Refinement of the Earth’s Gravity Field

Ning Jinsheng,Luo Zhicai,Chen Yongqi

Journal Article

Study on Down Knock-in Renewal Option Pricing

Wu Yun,He Jianmin

Journal Article

Numerical Computation of Flow and Pollutant Diffusion - Transportation With Non-orthogonal Curvilinear Coordinates

Wu Xiugang,Shen Yongming,Zheng Yonghong,Yang Zhifeng

Journal Article

The Ejection and Sweep in the Wall Region of a Turbulent Boundary Layer Under Adverse Pressure Gradient

Zhang Qiang,Lu Lipeng

Journal Article

Fractional-order global optimal backpropagation machine trained by an improved fractional-order steepest descent method

Yi-fei Pu, Jian Wang,puyifei@scu.edu.cn,wangjiannl@upc.edu.cn

Journal Article

Reinforcement design of anchor blocks for external tendons

Hong Hao,Tong Yuqiang ,He Zhiqi,Liu Zhao

Journal Article

The Nanoscale Density Gradient as a Structural Stabilizer for Glass Formation

Shaoxiong Zhou, Bangshao Dong, Yanguo Wang, Jingyu Qin, Weihua Wang

Journal Article

A Weighted Block-matching Criterion for the Hardware Implementation of Motion Estimators

Zhang Xia,Zheng Nanning,Zhang Guanglie,Wu Yong,Wang Shaorui,Xu Weipu

Journal Article

Survey on Extraction Methods of Transition Region

Liu Suolan,Yang Jingyu

Journal Article

Quantifying multiple social relationships based on a multiplex stochastic block model

Mincheng Wu, Zhen Li, Cunqi Shao, Shibo He,s18he@zju.edu.cn

Journal Article

A distributed stochastic optimization algorithm with gradient-tracking and distributed heavy-ball acceleration

Bihao Sun, Jinhui Hu, Dawen Xia, Huaqing Li,huaqingli@swu.edu.cn

Journal Article