Author

Yu Hong

1,153 papers · 1980–2026

1980peak 77 in 20102026

Frequent co-authors

Papers

showing 1–25 of 1,153 RSS

Convergence Properties of the BFGS Algoritm

Yu‐Hong Dai

Numerical Analysis · SIAM Journal on Optimization · 2002 · cited 328× · DOI ↗ · more like this ·

The BFGS method is one of the most famous quasi-Newton algorithms for unconstrained optimization. In 1984, Powell presented an example of a function of two variables that shows that the Polak--Ribière--Polyak (PRP) conjugate gradient method and the BFGS quasi-Newton method may cycle around eight nonstationary points if each line search picks a local minimum that provides a reduction in the objective function. In this paper, a new technique of choosing parameters is introduced, and an example with only six cyclic points is provided. It is also noted through the examples that the BFGS method with Wolfe line searches need not converge for nonconvex objective functions.

A Nonlinear Conjugate Gradient Algorithm with an Optimal Property and an Improved Wolfe Line Search

Yu‐Hong Dai, Caixia Kou

Numerical Analysis · SIAM Journal on Optimization · 2013 · cited 302× · DOI ↗ · more like this ·

In this paper, we seek the conjugate gradient direction closest to the direction of the scaled memoryless BFGS method and propose a family of conjugate gradient methods for unconstrained optimization. An improved Wolfe line search is also proposed, which can avoid a numerical drawback of the original Wolfe line search and guarantee the global convergence of the conjugate gradient method under mild conditions. To accelerate the algorithm, we introduce adaptive restarts along negative gradients based on the extent to which the function approximates some quadratic function during previous iterations. Numerical experiments with the CUTEr collection show that the proposed algorithm is promising.

Convergence Properties of Nonlinear Conjugate Gradient Methods

Yu‐Hong Dai, Jiye Han, Guanghui Liu +3 more

Numerical Analysis · SIAM Journal on Optimization · 2000 · cited 243× · DOI ↗ · more like this ·

Recently, important contributions on convergence studies of conjugate gradient methods were made by Gilbert and Nocedal [SIAM J. Optim., 2 (1992), pp. 21--42]. They introduce a "sufficient descent condition" to establish global convergence results. Although this condition is not needed in the convergence analyses of Newton and quasi-Newton methods, Gilbert and Nocedal hint that the sufficient descent condition, which was enforced by their two-stage line search algorithm, may be crucial for ensuring the global convergence of conjugate gradient methods. This paper shows that the sufficient descent condition is actually not needed in the convergence analyses of conjugate gradient methods. Consequently, convergence results on the Fletcher--Reeves- and Polak--Ribière-type methods are established in the absence of the sufficient descent condition. To show the differences between the…

Using Cross-Entity Inference to Improve Event Extraction

Yu Hong, Jianfeng Zhang, Bin Ma +3 more

Artificial Intelligence · Meeting of the Association for Computational Linguistics · 19 Jun 2011 · cited 231× · OpenAlex ↗ · more like this ·

Event extraction is the task of detecting certain specified types of events that are mentioned in the source language data. The state-of-the-art research on the task is transductive inference (e.g. cross-event inference). In this paper, we propose a new method of event extraction by well using cross-entity inference. In contrast to previous inference methods, we regard entitytype consistency as key feature to predict event mentions. We adopt this inference method to improve the traditional sentence-level event extraction system. Experiments show that we can get 8.6 % gain in trigger (event) identification, and more than 11.8 % gain for argument (role) classification in ACE event extraction. 1

Barzilai-Borwein Step Size for Stochastic Gradient Descent

Conghui Tan, Shiqian Ma, Yu-Hong Dai +1 more

math.OC · 23 May 2016 · cited 207× · arXiv ↗ · more like this ·

One of the major issues in stochastic gradient descent (SGD) methods is how to choose an appropriate step size while running the algorithm. Since the traditional line search technique does not apply for stochastic optimization algorithms, the common practice in SGD is either to use a diminishing step size, or to tune a fixed step size by hand, which can be time consuming in practice. In this paper, we propose to use the Barzilai-Borwein (BB) method to automatically compute step sizes for SGD and its variant: stochastic variance reduced gradient (SVRG) method, which leads to two algorithms: SGD-BB and SVRG-BB. We prove that SVRG-BB converges linearly for strongly convex objective functions. As a by-product, we prove the linear convergence result of SVRG with Option I proposed in [10], whose convergence result is missing in the literature. Numerical experiments on standard data sets show…

The cyclic Barzilai-–Borwein method for unconstrained optimization

Yu‐Hong Dai, William W. Hager, Klaus Schittkowski +1 more

Numerical Analysis · IMA Journal of Numerical Analysis · 25 Mar 2006 · cited 202× · DOI ↗ · more like this ·

In the cyclic Barzilai–Borwein (CBB) method, the same Barzilai–Borwein (BB) stepsize is reused for m consecutive iterations. It is proved that CBB is locally linearly convergent at a local minimizer with positive definite Hessian. Numerical evidence indicates that when m > n/2 ≥ 3, where n is the problem dimension, CBB is locally superlinearly convergent. In the special case m = 3 and n = 2, it is proved that the convergence rate is no better than linear, in general. An implementation of the CBB method, called adaptive cyclic Barzilai–Borwein (ACBB), combines a non-monotone line search and an adaptive choice for the cycle length m. In numerical experiments using the CUTEr test problem library, ACBB performs better than the existing BB gradient algorithm, while it is competitive with the well-known PRP+ conjugate gradient algorithm.

All Real Eigenvalues of Symmetric Tensors

Chun-Feng Cui, Yu-Hong Dai, Jiawang Nie

math.NA · March 2014 · cited 158× · arXiv ↗ · more like this ·

This paper studies how to compute all real eigenvalues of a symmetric tensor. As is well known, the largest or smallest eigenvalue can be found by solving a polynomial optimization problem, while the other middle eigenvalues can not. We propose a new approach for computing all real eigenvalues sequentially, from the largest to the smallest. It uses Jacobian SDP relaxations in polynomial optimization. We show that each eigenvalue can be computed by solving a finite hierarchy of semidefinite relaxations. Numerical experiments are presented to show how to do this.

Constructing Information Networks Using One Single Model

Qi Li, Heng Ji, Yu Hong +1 more

Artificial Intelligence · 2014 · cited 115× · DOI ↗ · more like this ·

In this paper, we propose a new frame-work that unifies the output of three infor-mation extraction (IE) tasks- entity men-tions, relations and events as an informa-tion network representation, and extracts all of them using one single joint model based on structured prediction. This novel formulation allows different parts of the information network fully interact with each other. For example, many rela-tions can now be considered as the re-sultant states of events. Our approach achieves substantial improvements over traditional pipelined approaches, and sig-nificantly advances state-of-the-art end-to-end event argument extraction. 1

Prediction of drug–target binding affinity using similarity-based convolutional neural network

Jooyong Shim, Zhen-Yu Hong, Insuk Sohn +1 more

Computational Theory and Mathematics · Scientific Reports · 24 Feb 2021 · cited 101× · DOI ↗ · more like this ·

Identifying novel drug-target interactions (DTIs) plays an important role in drug discovery. Most of the computational methods developed for predicting DTIs use binary classification, whose goal is to determine whether or not a drug-target (DT) pair interacts. However, it is more meaningful but also more challenging to predict the binding affinity that describes the strength of the interaction between a DT pair. If the binding affinity is not sufficiently large, such drug may not be useful. Therefore, the methods for predicting DT binding affinities are very valuable. The increase in novel public affinity data available in the DT-related databases enables advanced deep learning techniques to be used to predict binding affinities. In this paper, we propose a similarity-based model that applies 2-dimensional (2D) convolutional neural network (CNN) to the outer products between column…

HiFi4G: High-Fidelity Human Performance Rendering via Compact Gaussian Splatting

Yuheng Jiang, Zhehao Shen, Penghao Wang +5 more

cs.CV · 7 Dec 2023 · cited 98× · arXiv ↗ · more like this ·

We have recently seen tremendous progress in photo-real human modeling and rendering. Yet, efficiently rendering realistic human performance and integrating it into the rasterization pipeline remains challenging. In this paper, we present HiFi4G, an explicit and compact Gaussian-based approach for high-fidelity human performance rendering from dense footage. Our core intuition is to marry the 3D Gaussian representation with non-rigid tracking, achieving a compact and compression-friendly representation. We first propose a dual-graph mechanism to obtain motion priors, with a coarse deformation graph for effective initialization and a fine-grained Gaussian graph to enforce subsequent constraints. Then, we utilize a 4D Gaussian optimization scheme with adaptive spatial-temporal regularizers to effectively balance the non-rigid prior and Gaussian updating. We also present a companion…

Alternate step gradient method*

Yu‐Hong Dai

Numerical Analysis · Optimization · August 2003 · cited 92× · DOI ↗ · more like this ·

The Barzilai and Borwein (BB) gradient method does not guarantee a descent in the objective function at each iteration, but performs better than the classical steepest descent (SD) method in practice. So far, the BB method has found many successful applications and generalizations in linear systems, unconstrained optimization, convex-constrained optimization, stochastic optimization, etc. In this article, we propose a new gradient method that uses the SD and the BB steps alternately. Hence the name “alternate step (AS) gradient method.” Our theoretical and numerical analyses show that the AS method is a promising alternative to the BB method for linear systems. Unconstrained optimization algorithms related to the AS method are also discussed. Particularly, a more efficient gradient algorithm is provided by exploring the idea of the AS method in the GBB algorithm by Raydan (1997). To…

Fast high-resolution appearance editing using superimposed projections

Daniel G. Aliaga, Yu‐Hong Yeung, Alvin Law +2 more

Computer Vision and Pattern Recognition · ACM Transactions on Graphics · April 2012 · cited 89× · DOI ↗ · more like this ·

We present a system that superimposes multiple projections onto an object of arbitrary shape and color to produce high-resolution appearance changes. Our system produces appearances at an improved resolution compared to prior works and can change appearances at near interactive rates. Three main components are central to our system. First, the problem of computing compensation images is formulated as a constrained optimization which yields high-resolution appearances. Second, decomposition of the target appearance into base and scale images enables fast swapping of appearances on the object by requiring the constrained optimization to be computed only once per object. Finally, to make high-quality appearance edits practical, an elliptical Gaussian is used to model projector pixels and their interaction between projectors. To the best of our knowledge, we build the first system that…

Reform of State-owned Enterprises in China: The Chinese Communist Party Strikes Back

Yu Hong

Political Science and International Relations · Asian Studies Review · 3 Apr 2019 · cited 85× · DOI ↗ · more like this ·

State-owned enterprises (SOEs) play an important role in the Chinese economy and are also crucial instruments for implementing the Party-state’s policies and strategic initiatives. SOE reform is not only crucial to China’s transformation from an economy overly dependent on exports and investment to one more driven by domestic consumption, but is also a test of the political will and capability of the Chinese authorities to develop a market-based domestic economy. Despite the Chinese government’s commitment since 2013 to forging ahead with the SOE reform process, little real progress has been made so far. Reform of the SOEs is a central component of the government’s agenda for sustaining domestic economic growth and tightening the political control of the Chinese Communist Party over all aspects of the nation. The current wave of SOE reform has, however, focused on the state’s push for…

The Guangdong-Hong Kong-Macau Greater Bay Area in the making: development plan and challenges

Yu Hong

Political Science and International Relations · Cambridge Review of International Affairs · 23 Oct 2019 · cited 79× · DOI ↗ · more like this ·

The Chinese authority intends to develop the Guangdong-Hong Kong-Macau Greater Bay Area (GBA) as a new engine to sustain Chinese development. By applying the theories of state rescaling and city-cluster formation, this paper intends to conduct a case study of the GBA to examine the state-initiated plan for city-cluster formation in China, to analyse the Chinese central and local governments’ efforts to put this plan into action and the hurdles they face in formatting this city-cluster. The formulation of the GBA development plan for integrated development reflects a dynamic process of upscaling and downscaling of various levels of state power. The state’s plan for the GBA is to create a world-class city-cluster, and to eliminate barriers that hinder the free flow of goods and resources. Nevertheless, it will be challenging for Guangdong, Hong Kong and Macau, which fall under different…

Self-regulation: Employing a Generative Adversarial Network to Improve Event Detection

Yu Hong, Wenxuan Zhou, Jingli Zhang +2 more

Artificial Intelligence · 2018 · cited 67× · DOI ↗ · more like this ·

Due to the ability of encoding and mapping semantic information into a highdimensional latent feature space, neural networks have been successfully used for detecting events to a certain extent. However, such a feature space can be easily contaminated by spurious features inherent in event detection. In this paper, we propose a self-regulated learning approach by utilizing a generative adversarial network to generate spurious features. On the basis, we employ a recurrent network to eliminate the fakes. Detailed experiments on the ACE 2005 and TAC-KBP 2015 corpora show that our proposed method is highly effective and adaptable.

Joint power and admission control via p norm minimization deflation

Ya-Feng Liu, Yu-Hong Dai

cs.IT · 7 Mar 2013 · cited 65× · arXiv ↗ · more like this ·

In an interference network, joint power and admission control aims to support a maximum number of links at their specified signal to interference plus noise ratio (SINR) targets while using a minimum total transmission power. In our previous work, we formulated the joint control problem as a sparse $\ell_0$-minimization problem and relaxed it to a $\ell_1$-minimization problem. In this work, we propose to approximate the $\ell_0$-optimization problem to a p norm minimization problem where $0<p<1$, since intuitively p norm will approximate 0 norm better than 1 norm. We first show that the $\ell_p$-minimization problem is strongly NP-hard and then derive a reformulation of it such that the well developed interior-point algorithms can be applied to solve it. The solution to the $\ell_p$-minimization problem can efficiently guide the link's removals (deflation). Numerical simulations show…

Stabilized Barzilai-Borwein method

Oleg Burdakov, Yu-Hong Dai, Na Huang

math.OC · July 2019 · cited 63× · arXiv ↗ · more like this ·

The Barzilai-Borwein (BB) method is a popular and efficient tool for solving large-scale unconstrained optimization problems. Its search direction is the same as for the steepest descent (Cauchy) method, but its stepsize rule is different. Owing to this, it converges much faster than the Cauchy method. A feature of the BB method is that it may generate too long steps, which throw the iterates too far away from the solution. Moreover, it may not converge, even when the objective function is strongly convex. In this paper, a stabilization technique is introduced. It consists in bounding the distance between each pair of successive iterates, which often allows for decreasing the number of BB iterations. When the BB method does not converge, our simple modification of this method makes it convergent. Under suitable assumptions, we prove its global convergence, despite the fact that no line…

How to think about cyber sovereignty: the case of China

Yu Hong, G. Thomas Goodnight

Political Science and International Relations · Chinese Journal of Communication · 12 Nov 2019 · cited 62× · DOI ↗ · more like this ·

The cybersphere constitutes a global disagreement space where the contested, ongoing ties that link states and the internet come into being. In this paper, a critique of sovereignty and political economy is offered to evaluate contemporary controversies concerning authority, independence, regulation, and access to communications as matters of relationality, materiality, and disagreement. We review China’s promotion of cyber sovereignty as a complicating episode that reflects the development stresses of the sphere. China wishes to establish guardrails for the practices of multipolar global digital capitalism; yet, it has ushered in a variety of internet-related global issues, including in security, privacy, material well-being, developmental justice, and planetary futures. These complex aims invite and expand the dialectical spaces animating the cybersphere.

Analysis of monotone gradient methods

Yu‐Hong Dai, Ya-xiang Yuan

Numerical Analysis · Journal of Industrial and Management Optimization · 2005 · cited 61× · DOI ↗ · more like this ·

The gradient method is one simple method in nonlinear optimization.In this paper, we give a brief review on monotone gradient methods andstudy their numerical properties by introducing a new technique oflong-term observation. We find that, one monotone gradient algorithmwhich is proposed by Yuan recently shares with the Barzilai-Borwein (BB)method the property that the gradientcomponents with respect to the eigenvectors of the function Hessian aredecreasing together. This might partly explain why this algorithm by Yuanis comparable to the BB method in practice. Some examples are alsoprovided showing that the alternate minimization algorithm and theother algorithm by Yuan may fall into cycles. Some more efficientgradient algorithms are provided. Particularly, one of them is monotoneand performs better than the BB method in the quadratic case.

China's Belt and Road Initiative and Its Implications for Southeast Asia

Yu Hong

Political Science and International Relations · Asia policy · July 2017 · cited 56× · DOI ↗ · more like this ·

China's Belt and Road Initiative and Its Implications for Southeast Asia Hong Yu (bio) During state visits to Kazakhstan and Indonesia in 2013, Xi Jinping announced the land-based Silk Road Economic Belt and the sea-based 21st Century Maritime Silk Road, respectively. Shortly after that, these two initiatives were combined to form one unified concept, known as the Belt and Road Initiative (BRI). This grand initiative, comprising various routes by sea and land, is intended to connect China with Southeast and South Asia, Central Asia, Pacific Oceania, Africa, and Europe. BRI is centered on both soft and hard infrastructure connectivity, aiming to forge an integrated and extensive network of regional infrastructure with China at its hub. BRI has gradually emerged as a top Chinese national strategy. Given China's emergence as a global power through industrial redeployment and outward…

Complete and faithful hyperentangled-Bell-state analysis of photon systems using a failure-heralded and fidelity-robust quantum gate

Cong Cao, Li Zhang, Yu‐Hong Han +4 more

Artificial Intelligence · Optics Express · 11 Jan 2020 · cited 51× · DOI ↗ · more like this ·

Hyperentangled-Bell-state analysis (HBSA) represents a key step in many quantum information processing schemes that utilize hyperentangled states. In this paper, we present a complete and faithful HBSA scheme for two-photon quantum systems hyperentangled in both the polarization and spatial-mode degrees of freedom, using a failure-heralded and fidelity-robust quantum swap gate for the polarization states of two photons (P-SWAP gate), constructed with a singly charged semiconductor quantum dot (QD) in a double-sided optical microcavity (double-sided QD-cavity system) and some linear-optical elements. Compared with the previously proposed complete HBSA schemes using different auxiliary tools such as parity-check quantum nondemonlition detectors or additional entangled states, our scheme significantly simplifies the analysis process and saves the quantum resource. Unlike the previous…

Robust Dual Gaussian Splatting for Immersive Human-centric Volumetric Videos

Yuheng Jiang, Zhehao Shen, Yu Hong +5 more

cs.GR · 12 Sep 2024 · cited 50× · arXiv ↗ · more like this ·

Volumetric video represents a transformative advancement in visual media, enabling users to freely navigate immersive virtual experiences and narrowing the gap between digital and real worlds. However, the need for extensive manual intervention to stabilize mesh sequences and the generation of excessively large assets in existing workflows impedes broader adoption. In this paper, we present a novel Gaussian-based approach, dubbed \textit{DualGS}, for real-time and high-fidelity playback of complex human performance with excellent compression ratios. Our key idea in DualGS is to separately represent motion and appearance using the corresponding skin and joint Gaussians. Such an explicit disentanglement can significantly reduce motion redundancy and enhance temporal coherence. We begin by initializing the DualGS and anchoring skin Gaussians to joint Gaussians at the first frame.…

Applications of Quantum Chemistry in Pharmaceutical Process Development: Current State and Opportunities

Yu‐hong Lam, Yuriy A. Abramov, Ravi S. Ananthula +8 more

Computational Theory and Mathematics · Organic Process Research & Development · 23 Jul 2020 · cited 49× · DOI ↗ · more like this ·

Application of computational methods to understanding and predicting properties of analogues for drug discovery has enjoyed a long history of success. However, the drug development space (post-candidate selection) is currently experiencing a rapid growth in this arena. Due to the revolution in computing hardware development and improved computational techniques, quantum chemical (QC) calculations have become an essential tool in this space, allowing results from complex calculations to inform chemical development efforts. As a result, numerous pharmaceutical companies are employing QC as part of their drug development workflow. Calculations cover the range of transition state calculations, reaction path determination, and potential energy surface scans, among others. The impact of this rapid growth is realized by providing an in-depth understanding of chemical processes and predictive…

Enhancement of Image Contrast, Stability, and SALDI-MS Detection Sensitivity for Latent Fingerprint Analysis by Tuning the Composition of Silver–Gold Nanoalloys

Yu-Hong Cheng, Yue Zhang, Siu-Leung Chau +3 more

Safety Research · ACS Applied Materials & Interfaces · 17 Oct 2016 · cited 49× · DOI ↗ · more like this ·

Metal alloy nanoparticles (NPs) offer a new combination of unique physicochemical properties based on their pure counterparts, which can facilitate the development of novel analytical methods. Here, we demonstrated that Ag-Au alloy NPs could be utilized for optical and mass spectrometric imaging of latent fingerprints (LFPs) with improved image contrast, stability, and detection sensitivity. Upon deposition of Ag-Au alloy NPs (Ag:Au = 60:40 wt %), ridge regions of the LFP became amber colored, while the groove regions appeared purple-blue. The presence of Au in the Ag-Au alloy NPs suppressed aggregation behavior compared to pure AgNPs, thus improving the stability of the developed LFP images. In addition, the Ag component in the Ag-Au alloy NPs enhanced optical absorption efficiency compared to pure AuNPs, resulting in higher contrast LFP images. Moreover, varying the Ag-Au ratio could…

Positive Semidefinite Generalized Diffusion Tensor Imaging via Quadratic Semidefinite Programming

Yannan Chen, Yu‐Hong Dai, Deren Han +1 more

Computational Mathematics · SIAM Journal on Imaging Sciences · 2013 · cited 47× · DOI ↗ · more like this ·

The positive definiteness of a diffusion tensor is important in magnetic resonance imaging because it reflects the phenomenon of water molecular diffusion in complicated biological tissue environments. To preserve this property, we represent it as an explicit positive semidefinite (PSD) matrix constraint and some linear matrix equalities. The objective function is the regularized linear least squares fitting for the log-linearized Stejskal--Tanner equation. The regularization term is the heuristic nuclear norm of the PSD matrix, since we expect it to be of low rank. In this way, we establish a convex quadratic semidefinite programming (SDP) model, whose global solution exists. The optimal solution could be solved by three efficient methods. While there are two state-of-the-art solvers---SDPT3 and QSDP---for the primal problem, we design a new augmented Lagrangian based alternating…