Hu Shao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Hu Shao has a total of 125 co-authors. Here are the top co-authors by number of shared publications:
- Pengjie Liu
School of Mathematics, China University of Mining and Technology, Xuzhou, China
17 shared publications Pengjie Liu
9 shared publicationsFeng Shao
7 shared publications- Ting Wu
School of Mathematics, Nanjing University, Nanjing, China
6 shared publications - Xiaoyu Wu
School of Mathematics, China University of Mining and Technology, Xuzhou, China
6 shared publications - Liang Shen
School of Management, Xuzhou Medical University, Xuzhou, China
5 shared publications
Latest Publications
- 13 November 2026
- Published by Elsevier in Applied Mathematical Modelling
- Vol. 164
Two extended conjugate gradient methods with adaptive restart for unconstrained optimization
- 13 November 2026
- Published by Elsevier in Applied Mathematical Modelling
- Vol. 164
Two extended conjugate gradient methods with adaptive restart for unconstrained optimization
- 11 November 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 335
Two-Stage traffic sensor location under uncertainty with cycle-Breaking screening
- 11 November 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 335
Two-Stage traffic sensor location under uncertainty with cycle-Breaking screening
- Open Access
- 26 August 2026
- Published by Elsevier in SSRN Electronic Journal
Two-Stage Traffic Sensor Location under Uncertainty with Cycle-Breaking Screening
The traffic sensor location problem determines where sensors should be installed so that full link flow observability can be achieved while sensing quality remains reliable under uncertainty. The problem becomes more difficult in cyclic networks, where local flow inference may fail because of unresolved cycles, and under uncertain sensing conditions, where sensor type and error exposure affect allocation decisions. This paper addresses the traffic sensor location problem in cyclic networks through a two-stage framework. Stage~1 selects observed links with a cycle-breaking model. Stage~2 allocates sensor types on those links by distributionally robust optimization (DRO) and sample average approximation (SAA) using error-propagation scores. The numerical study moves from a 12-link cyclic testbed to a medium-scale Eastern Massachusetts case and the large Anaheim network. On the 12-link testbed, Stage~1 attains the rank-based lower bound of 8 observed links, whereas one-third of node-priority orders under order-sensitive inference require a ninth sensor. On the 254-link Eastern case, IFOS selects 200 observed links, preserves an invertible unobserved subsystem, and yields lower noisy same-sparsity reconstruction error than greedy-rank. On Anaheim, the global inverse-based score becomes uniform, whereas a localized score differentiates links and changes upgrade priorities. Taken together, the results support a unified two-stage framework that connects cycle-breaking screening with uncertainty-aware sensor allocation in cyclic traffic networks.
- 26 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessTwo-Stage Traffic Sensor Location under Uncertainty with Cycle-Breaking Screening
The traffic sensor location problem determines where sensors should be installed so that full link flow observability can be achieved while sensing quality remains reliable under uncertainty. The problem becomes more difficult in cyclic networks, where local flow inference may fail because of unresolved cycles, and under uncertain sensing conditions, where sensor type and error exposure affect allocation decisions. This paper addresses the traffic sensor location problem in cyclic networks through a two-stage framework. Stage~1 selects observed links with a cycle-breaking model. Stage~2 allocates sensor types on those links by distributionally robust optimization (DRO) and sample average approximation (SAA) using error-propagation scores. The numerical study moves from a 12-link cyclic testbed to a medium-scale Eastern Massachusetts case and the large Anaheim network. On the 12-link testbed, Stage~1 attains the rank-based lower bound of 8 observed links, whereas one-third of node-priority orders under order-sensitive inference require a ninth sensor. On the 254-link Eastern case, IFOS selects 200 observed links, preserves an invertible unobserved subsystem, and yields lower noisy same-sparsity reconstruction error than greedy-rank. On Anaheim, the global inverse-based score becomes uniform, whereas a localized score differentiates links and changes upgrade priorities. Taken together, the results support a unified two-stage framework that connects cycle-breaking screening with uncertainty-aware sensor allocation in cyclic traffic networks.
- 18 August 2026
- Published by Global Science Press in Numerical Mathematics: Theory, Methods and Applications
- Vol. 19 (3) , 603-644
Stochastic ADMM with Balanced Augmented Lagrangian Method for Nonconvex and Nonsmooth Finite-Sum Optimization
In this paper, we propose a balanced augmented Lagrangian method based on accelerated stochastic ADMM (b-ASADMM) to efficiently solve structured separable nonconvex optimization problems subject to linear constraints. The objective function in this problem comprises potentially nonsmooth and smooth functions, where the smooth function is an average of multiple nonconvex smooth functions. The involved smooth subproblem is tackled by an accelerated stochastic gradient method based on weighting of stochastic item and pre-variable. The involved nonsmooth subproblem is solved under incorporation of Bregman distance to avoid the case that subproblem does not have a closed-form solution due to the complicated quadratic term or other hindering. The involved balanced augmented Lagrangian method advances the original ALM by balancing its subproblems and improving its implementation. In contrast to most deterministic and stochastic ADMMs, our dual variable allows a more flexible and larger step-size region. By standard smoothness assumption, we establish the global convergence and iteration complexity of the generated sequence. Furthermore, we provide a linear convergence rate of b-ASADMM under a local error bound condition and the weakly convex property of the nonsmooth component. Numerical experiments on the graph-guided fused Lasso problem and the smooth clipped absolute deviation penalty problem are conducted to verify the effectiveness of b-ASADMM.
- 18 August 2026
- Published by Global Science Press in Numerical Mathematics: Theory, Methods and Applications
- Vol. 19 (3) , 603-644
Stochastic ADMM with Balanced Augmented Lagrangian Method for Nonconvex and Nonsmooth Finite-Sum Optimization
In this paper, we propose a balanced augmented Lagrangian method based on accelerated stochastic ADMM (b-ASADMM) to efficiently solve structured separable nonconvex optimization problems subject to linear constraints. The objective function in this problem comprises potentially nonsmooth and smooth functions, where the smooth function is an average of multiple nonconvex smooth functions. The involved smooth subproblem is tackled by an accelerated stochastic gradient method based on weighting of stochastic item and pre-variable. The involved nonsmooth subproblem is solved under incorporation of Bregman distance to avoid the case that subproblem does not have a closed-form solution due to the complicated quadratic term or other hindering. The involved balanced augmented Lagrangian method advances the original ALM by balancing its subproblems and improving its implementation. In contrast to most deterministic and stochastic ADMMs, our dual variable allows a more flexible and larger step-size region. By standard smoothness assumption, we establish the global convergence and iteration complexity of the generated sequence. Furthermore, we provide a linear convergence rate of b-ASADMM under a local error bound condition and the weakly convex property of the nonsmooth component. Numerical experiments on the graph-guided fused Lasso problem and the smooth clipped absolute deviation penalty problem are conducted to verify the effectiveness of b-ASADMM.
- 9 February 2026
- Published by Springer Nature in Journal of Applied Mathematics and Computing
- Vol. 72 (3) , 128
An adaptive restart Riemannian spectral Dai-Kou conjugate gradient method with quasi-Newton update
- 9 February 2026
- Published by Springer Nature in Journal of Applied Mathematics and Computing
- Vol. 72 (3) , 128
An adaptive restart Riemannian spectral Dai-Kou conjugate gradient method with quasi-Newton update
- 9 February 2026
- Published by Elsevier in Journal of Computational and Applied Mathematics
- Vol. 483
A θ-inertial ADMM for efficient and stable nonconvex optimization
- 9 February 2026
- Published by Elsevier in Journal of Computational and Applied Mathematics
- Vol. 483
A θ-inertial ADMM for efficient and stable nonconvex optimization