Qingsong Wang

Dr. Qingsong Wang


Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 March 2023 in Journal of Optimization Theory and Applications
Verified
Journal Article
Closed access
PDF file
QWQingsong Wang
Qingsong Wang
CCChunfeng Cui
Chunfeng Cui
DHDeren Han
Deren Han

24 March 2024 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Open access
PDF file
QWQingsong Wang
Qingsong Wang
ZLZehui Liu
Zehui Liu
CCChunfeng Cui
Chunfeng Cui
DHDeren Han
Deren Han
In this paper, we explore a specific optimization problem that involves the combination of a differentiable nonconvex function and a nondifferentiable function. The differentiable component lacks a global Lipschitz continuous gradient, posing challenges for optimization. To address this issue and accelerate the convergence, we propose a Bregman proximal stochastic gradient method with extrapolation (BPSGE), which only requires smooth adaptivity of the differentiable part. Under variance reduction framework, we not only analyze the subsequential and global convergence of the proposed algorithm under certain conditions, but also analyze the sublinear convergence rate of the subsequence, and the complexity of the algorithm, revealing that the BPSGE algorithm requires at most O(epsilon\^\,(-2)) iterations in expectation to attain an epsilon-stationary point. To validate the effectiveness of our proposed algorithm, we conduct numerical experiments on three real-world applications: graph regularized nonnegative matrix factorization (NMF), matrix factorization with weakly-convex regularization, and NMF with nonconvex sparsity constraints. These experiments demonstrate that BPSGE is faster than the baselines without extrapolation. The code is available at: https://github.com/nothing2wang/BPSGE-Algorithm.

7 October 2024 in IEEE Signal Processing Letters
Verified
Journal Article
Closed access
QWQingsong Wang
Qingsong Wang
CCChunfeng Cui
Chunfeng Cui
DHDeren Han
Deren Han
Recently, there has been a growing interest in the exploration of Nonlinear Matrix Decomposition (NMD) due to its close ties with neural networks. NMD aims to find a low-rank matrix from a sparse nonnegative matrix with a per-element nonlinear function. A typical choice is the Rectified Linear Unit (ReLU) activation function. To address over-fitting in the existing ReLU-based NMD model (ReLU-NMD), we propose a Tikhonov regularized ReLU-NMD model, referred to as ReLU-NMD-T. Subsequently, we introduce a momentum accelerated algorithm for handling the ReLU-NMD-T model. A distinctive feature, setting our work apart from most existing studies, is the incorporation of both positive and negative momentum parameters in our algorithm. Our numerical experiments on real-world datasets show the effectiveness of the proposed model and algorithm. Moreover, the code is available at https://github.com/nothing2wang/NMD-TM .

17 February 2025 in Computational Optimization and Applications
Verified
Journal Article
Closed access
PDF file
ZLZehui Liu
Zehui Liu
QWQingsong Wang
Qingsong Wang
CCChunfeng Cui
Chunfeng Cui
YXYong Xia
Yong Xia

14 November 2022 in IEEE Transactions on Signal Processing
Verified
Journal Article
Closed access
DNDunbiao Niu
Dunbiao Niu
CWChengjing Wang
Chengjing Wang
PTPeipei Tang
Peipei Tang
QWQingsong Wang
Qingsong Wang
ESEnbin Song
Enbin Song
Support vector machines (SVMs) are successful supervised learning models that analyze data for classification and regression. Previous work has demonstrated the superiority of the SVMs in dealing with the high dimensional, low sample size problems. However, the increase of the sample size brings great challenges to accurately and efficiently solve the large-scale SVMs, especially for the nonlinear kernel SVMs, which may lead to huge computational costs and unaffordable storage burden. In this paper, we propose a highly efficient sparse semismooth Newton (SsN) based augmented Lagrangian (AL) method for solving a class of large-scale SVMs that can be formulated as a convex quadratic programming problem with a linear equality constraint and a simple box constraint. The asymptotic superlinear convergence rate of both the primal and the dual iteration sequences generated by the AL method is guaranteed due to the piecewise linear-quadratic structure of the problem. Furthermore, we reveal the close connection between the number of support vectors and the sparse structure of the generalized Jacobian for the inner subproblem of the AL method. By exploiting this hidden sparsity, the inner subproblem can be solved by the SsN method efficiently and accurately, which greatly reduces the storage burden and computational costs. In particular, for the nonlinear kernel SVMs, since the sparse structure may not manifest in the early iterations of the AL method, we solve a linear kernel SVM approximated by the random Fourier features method to produce a good initial point, and then transfer to solve the original problem. Numerical experiments demonstrate that the proposed algorithm outperforms the current state-of-the-art solvers for the large-scale SVMs.

6 February 2024 in Mathematical Methods of Operations Research
Verified
Journal Article
Closed access
PDF file
QWQingsong Wang
Qingsong Wang
DHDeren Han
Deren Han