5 Search Results for "Cai, Shaowei"

Document
DOI: 10.4230/LIPIcs.CP.2023.12
Towards More Efficient Local Search for Pseudo-Boolean Optimization

Authors: Yi Chu, Shaowei Cai, Chuan Luo, Zhendong Lei, and Cong Peng

Published in: LIPIcs, Volume 280, 29th International Conference on Principles and Practice of Constraint Programming (CP 2023)

Abstract
Pseudo-Boolean (PB) constraints are highly expressive, and many combinatorial optimization problems can be modeled using pseudo-Boolean optimization (PBO). It is recognized that stochastic local search (SLS) is a powerful paradigm for solving combinatorial optimization problems, but the development of SLS for solving PBO is still in its infancy. In this paper, we develop an effective SLS algorithm for solving PBO, dubbed NuPBO, which introduces a novel scoring function for PB constraints and a new weighting scheme. We conduct experiments on a broad range of six public benchmarks, including three real-world benchmarks, a benchmark from PB competition, an integer linear programming optimization benchmark, and a crafted combinatorial benchmark, to compare NuPBO against five state-of-the-art competitors, including a recently-proposed SLS PBO solver LS-PBO, two complete PB solvers PBO-IHS and RoundingSat, and two mixed integer programming (MIP) solvers Gurobi and SCIP. NuPBO has been exhibited to perform best on these three real-world benchmarks. On the other three benchmarks, NuPBO shows competitive performance compared to state-of-the-art competitors, and it significantly outperforms LS-PBO, indicating that NuPBO greatly advances the state of the art in SLS for solving PBO.
Cite as

Yi Chu, Shaowei Cai, Chuan Luo, Zhendong Lei, and Cong Peng. Towards More Efficient Local Search for Pseudo-Boolean Optimization. In 29th International Conference on Principles and Practice of Constraint Programming (CP 2023). Leibniz International Proceedings in Informatics (LIPIcs), Volume 280, pp. 12:1-12:18, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2023)

Copy BibTex To Clipboard

@InProceedings{chu_et_al:LIPIcs.CP.2023.12,
  author =	{Chu, Yi and Cai, Shaowei and Luo, Chuan and Lei, Zhendong and Peng, Cong},
  title =	{{Towards More Efficient Local Search for Pseudo-Boolean Optimization}},
  booktitle =	{29th International Conference on Principles and Practice of Constraint Programming (CP 2023)},
  pages =	{12:1--12:18},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-300-3},
  ISSN =	{1868-8969},
  year =	{2023},
  volume =	{280},
  editor =	{Yap, Roland H. C.},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops-dev.dagstuhl.de/entities/document/10.4230/LIPIcs.CP.2023.12},
  URN =		{urn:nbn:de:0030-drops-190490},
  doi =		{10.4230/LIPIcs.CP.2023.12},
  annote =	{Keywords: Pseudo-Boolean Optimization, Stochastic Local Search, Scoring Function, Weighting Scheme}
}
Document
DOI: 10.4230/LIPIcs.CP.2023.41
Improving Local Search for Pseudo Boolean Optimization by Fragile Scoring Function and Deep Optimization

Authors: Wenbo Zhou, Yujiao Zhao, Yiyuan Wang, Shaowei Cai, Shimao Wang, Xinyu Wang, and Minghao Yin

Published in: LIPIcs, Volume 280, 29th International Conference on Principles and Practice of Constraint Programming (CP 2023)

Abstract
Pseudo-Boolean optimization (PBO) is usually used to model combinatorial optimization problems, especially for some real-world applications. Despite its significant importance in both theory and applications, there are few works on using local search to solve PBO. This paper develops a novel local search framework for PBO, which has three main ideas. First, we design a two-level selection strategy to evaluate all candidate variables. Second, we propose a novel deep optimization strategy to disturb some search spaces. Third, a sampling flipping method is applied to help the algorithm jump out of local optimum. Experimental results show that the proposed algorithms outperform three state-of-the-art PBO algorithms on most instances.
Cite as

Wenbo Zhou, Yujiao Zhao, Yiyuan Wang, Shaowei Cai, Shimao Wang, Xinyu Wang, and Minghao Yin. Improving Local Search for Pseudo Boolean Optimization by Fragile Scoring Function and Deep Optimization. In 29th International Conference on Principles and Practice of Constraint Programming (CP 2023). Leibniz International Proceedings in Informatics (LIPIcs), Volume 280, pp. 41:1-41:18, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2023)

Copy BibTex To Clipboard

@InProceedings{zhou_et_al:LIPIcs.CP.2023.41,
  author =	{Zhou, Wenbo and Zhao, Yujiao and Wang, Yiyuan and Cai, Shaowei and Wang, Shimao and Wang, Xinyu and Yin, Minghao},
  title =	{{Improving Local Search for Pseudo Boolean Optimization by Fragile Scoring Function and Deep Optimization}},
  booktitle =	{29th International Conference on Principles and Practice of Constraint Programming (CP 2023)},
  pages =	{41:1--41:18},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-300-3},
  ISSN =	{1868-8969},
  year =	{2023},
  volume =	{280},
  editor =	{Yap, Roland H. C.},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops-dev.dagstuhl.de/entities/document/10.4230/LIPIcs.CP.2023.41},
  URN =		{urn:nbn:de:0030-drops-190784},
  doi =		{10.4230/LIPIcs.CP.2023.41},
  annote =	{Keywords: Local Search, Pseudo-Boolean Optimization, Deep Optimization}
}
Document
Short Paper
DOI: 10.4230/LIPIcs.CP.2021.5
Improving Local Search for Structured SAT Formulas via Unit Propagation Based Construct and Cut Initialization (Short Paper)

Authors: Shaowei Cai, Chuan Luo, Xindi Zhang, and Jian Zhang

Published in: LIPIcs, Volume 210, 27th International Conference on Principles and Practice of Constraint Programming (CP 2021)

Abstract
This work is dedicated to improving local search solvers for the Boolean satisfiability (SAT) problem on structured instances. We propose a construct-and-cut (CnC) algorithm based on unit propagation, which is used to produce initial assignments for local search. We integrate our CnC initialization procedure within several state-of-the-art local search SAT solvers, and obtain the improved solvers. Experiments are carried out with a benchmark encoded from a spectrum repacking project as well as benchmarks encoded from two important mathematical problems namely Boolean Pythagorean Triple and Schur Number Five. The experiments show that the CnC initialization improves the local search solvers, leading to better performance than state-of-the-art SAT solvers based on Conflict Driven Clause Learning (CDCL) solvers.
Cite as

Shaowei Cai, Chuan Luo, Xindi Zhang, and Jian Zhang. Improving Local Search for Structured SAT Formulas via Unit Propagation Based Construct and Cut Initialization (Short Paper). In 27th International Conference on Principles and Practice of Constraint Programming (CP 2021). Leibniz International Proceedings in Informatics (LIPIcs), Volume 210, pp. 5:1-5:10, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2021)

Copy BibTex To Clipboard

@InProceedings{cai_et_al:LIPIcs.CP.2021.5,
  author =	{Cai, Shaowei and Luo, Chuan and Zhang, Xindi and Zhang, Jian},
  title =	{{Improving Local Search for Structured SAT Formulas via Unit Propagation Based Construct and Cut Initialization}},
  booktitle =	{27th International Conference on Principles and Practice of Constraint Programming (CP 2021)},
  pages =	{5:1--5:10},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-211-2},
  ISSN =	{1868-8969},
  year =	{2021},
  volume =	{210},
  editor =	{Michel, Laurent D.},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops-dev.dagstuhl.de/entities/document/10.4230/LIPIcs.CP.2021.5},
  URN =		{urn:nbn:de:0030-drops-152969},
  doi =		{10.4230/LIPIcs.CP.2021.5},
  annote =	{Keywords: Satisfiability, Local Search, Unit Propagation, Mathematical Problems}
}
Document
DOI: 10.4230/LIPIcs.CP.2021.39
Improving Local Search for Minimum Weighted Connected Dominating Set Problem by Inner-Layer Local Search

Authors: Bohan Li, Kai Wang, Yiyuan Wang, and Shaowei Cai

Published in: LIPIcs, Volume 210, 27th International Conference on Principles and Practice of Constraint Programming (CP 2021)

Abstract
The minimum weighted connected dominating set (MWCDS) problem is an important variant of connected dominating set problems with wide applications, especially in heterogenous networks and gene regulatory networks. In the paper, we develop a nested local search algorithm called NestedLS for solving MWCDS on classic benchmarks and massive graphs. In this local search framework, we propose two novel ideas to make it effective by utilizing previous search information. First, we design the restart based smoothing mechanism as a diversification method to escape from local optimal. Second, we propose a novel inner-layer local search method to enlarge the candidate removal set, which can be modelled as an optimized version of spanning tree problem. Moreover, inner-layer local search method is a general method for maintaining the connectivity constraint when dealing with massive graphs. Experimental results show that NestedLS outperforms state-of-the-art meta-heuristic algorithms on most instances.
Cite as

Bohan Li, Kai Wang, Yiyuan Wang, and Shaowei Cai. Improving Local Search for Minimum Weighted Connected Dominating Set Problem by Inner-Layer Local Search. In 27th International Conference on Principles and Practice of Constraint Programming (CP 2021). Leibniz International Proceedings in Informatics (LIPIcs), Volume 210, pp. 39:1-39:16, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2021)

Copy BibTex To Clipboard

@InProceedings{li_et_al:LIPIcs.CP.2021.39,
  author =	{Li, Bohan and Wang, Kai and Wang, Yiyuan and Cai, Shaowei},
  title =	{{Improving Local Search for Minimum Weighted Connected Dominating Set Problem by Inner-Layer Local Search}},
  booktitle =	{27th International Conference on Principles and Practice of Constraint Programming (CP 2021)},
  pages =	{39:1--39:16},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-211-2},
  ISSN =	{1868-8969},
  year =	{2021},
  volume =	{210},
  editor =	{Michel, Laurent D.},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops-dev.dagstuhl.de/entities/document/10.4230/LIPIcs.CP.2021.39},
  URN =		{urn:nbn:de:0030-drops-153304},
  doi =		{10.4230/LIPIcs.CP.2021.39},
  annote =	{Keywords: Operations Research, NP-hard Problem, Local Search, Weighted Connected Dominating Set Problem}
}
Document
DOI: 10.4230/LIPIcs.SEA.2020.8
Effect of Initial Assignment on Local Search Performance for Max Sat

Authors: Daniel Berend and Yochai Twitto

Published in: LIPIcs, Volume 160, 18th International Symposium on Experimental Algorithms (SEA 2020)

Abstract
In this paper, we explore the correlation between the quality of initial assignments provided to local search heuristics and that of the corresponding final assignments. We restrict our attention to the Max r-Sat problem and to one of the leading local search heuristics - Configuration Checking Local Search (CCLS). We use a tailored version of the Method of Conditional Expectations (MOCE) to generate initial assignments of diverse quality. We show that the correlation in question is significant and long-lasting. Namely, even when we delve deeper into the local search, we are still in the shadow of the initial assignment. Thus, under practical time constraints, the quality of the initial assignment is crucial to the performance of local search heuristics. To demonstrate our point, we improve CCLS by combining it with MOCE. Instead of starting CCLS from random initial assignments, we start it from excellent initial assignments, provided by MOCE. Indeed, it turns out that this kind of initialization provides a significant improvement of this state-of-the-art solver. This improvement becomes more and more significant as the instance grows.
Cite as

Daniel Berend and Yochai Twitto. Effect of Initial Assignment on Local Search Performance for Max Sat. In 18th International Symposium on Experimental Algorithms (SEA 2020). Leibniz International Proceedings in Informatics (LIPIcs), Volume 160, pp. 8:1-8:14, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2020)

Copy BibTex To Clipboard

@InProceedings{berend_et_al:LIPIcs.SEA.2020.8,
  author =	{Berend, Daniel and Twitto, Yochai},
  title =	{{Effect of Initial Assignment on Local Search Performance for Max Sat}},
  booktitle =	{18th International Symposium on Experimental Algorithms (SEA 2020)},
  pages =	{8:1--8:14},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-148-1},
  ISSN =	{1868-8969},
  year =	{2020},
  volume =	{160},
  editor =	{Faro, Simone and Cantone, Domenico},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops-dev.dagstuhl.de/entities/document/10.4230/LIPIcs.SEA.2020.8},
  URN =		{urn:nbn:de:0030-drops-120823},
  doi =		{10.4230/LIPIcs.SEA.2020.8},
  annote =	{Keywords: Combinatorial Optimization, Maximum Satisfiability, Local Search, Probabilistic Algorithms}
}
  • Refine by Author
  • 4 Cai, Shaowei
  • 2 Luo, Chuan
  • 2 Wang, Yiyuan
  • 1 Berend, Daniel
  • 1 Chu, Yi
  • Show More...

  • Refine by Classification
  • 3 Theory of computation → Randomized local search
  • 1 Applied computing → Operations research
  • 1 Computing methodologies → Search methodologies
  • 1 Theory of computation → Stochastic approximation
  • 1 Theory of computation → Theory of randomized search heuristics

  • Refine by Keyword
  • 4 Local Search
  • 2 Pseudo-Boolean Optimization
  • 1 Combinatorial Optimization
  • 1 Deep Optimization
  • 1 Mathematical Problems
  • Show More...

  • Refine by Type
  • 5 document

  • Refine by Publication Year
  • 2 2021
  • 2 2023
  • 1 2020

Questions / Remarks / Feedback
X

Feedback for Dagstuhl Publishing


Thanks for your feedback!

Feedback submitted

Could not send message

Please try again later or send an E-mail
© 2023-2024 Schloss Dagstuhl – LZI GmbH Imprint Privacy Contact