{"type":"ScholarlyArticle","title":"BandMaxSAT: A Local Search MaxSAT Solver with Multi-armed Bandit","authors":[{"name":"Jiongzhi Zheng","slug":"jiongzhi-zheng","given":"Jiongzhi","family":"Zheng","affiliation":"School of Computer Science and Technology , Huazhong University of Science and Technology; Institute of Artificial Intelligence , Huazhong University of Science and Technology , China"},{"name":"Kun He","slug":"kun-he","given":"Kun","family":"He","affiliation":"School of Computer Science and Technology , Huazhong University of Science and Technology"},{"name":"Jianrong Zhou","slug":"jianrong-zhou","given":"Jianrong","family":"Zhou","affiliation":"School of Computer Science and Technology , Huazhong University of Science and Technology"},{"name":"Yan Jin","slug":"yan-jin","given":"Yan","family":"Jin","affiliation":"School of Computer Science and Technology , Huazhong University of Science and Technology"},{"name":"Chu-Min Li","slug":"chu-min-li","given":"Chu-Min","family":"Li","affiliation":"MIS , University of Picardie Jules Verne , France"},{"name":"Felip Manyà","slug":"felip-manya","given":"Felip","family":"Manyà","affiliation":"Artificial Intelligence Research Institute (IIIA) , CSIC , Bellaterra , Spain"}],"abstract":"We address Partial MaxSAT (PMS) and Weighted PMS (WPMS), two practical generalizations of the MaxSAT problem, and propose a local search algorithm called BandMaxSAT, that applies a multi-armed bandit to guide the search direction, for these problems. The bandit in our method is associated with all the soft clauses in the input (W)PMS instance. Each arm corresponds to a soft clause. The bandit model can help BandMaxSAT to select a good direction to escape from local optima by selecting a soft clause to be satisfied in the current step, that is, selecting an arm to be pulled. We further propose an initialization method for (W)PMS that prioritizes both unit and binary clauses when producing the initial solutions. Extensive experiments demonstrate that BandMaxSAT significantly outperforms the state-of-the-art (W)PMS local search algorithm SATLike3.0. Specifically, the number of instances in which BandMaxSAT obtains better results is about twice that obtained by SATLike3.0. We further combine BandMaxSAT with the complete solver TT-Open-WBO-Inc. The resulting solver BandMaxSAT-c also outperforms some of the best state-of-the-art complete (W)PMS solvers, including SATLike-c, Loandra and TT-Open-WBO-Inc.","doi":"10.24963/ijcai.2022/264","url":"https://ijcai.org/proceedings/2022/264","pdf":"https://ijcai.org/proceedings/2022/0264.pdf","markdown":"https://ijcai.org/proceedings/2022/264.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/264","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"Constraint Satisfaction and Optimization","track":"Main Track","number":264,"pages":{"first":1901,"last":1907},"keywords":["Constraint Satisfaction and Optimization: Satisfiabilty","Constraint Satisfaction and Optimization: Constraint Optimization"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38986621","length_sec":88,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38984764","length_sec":359,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Ansótegui et al., 2022] Carlos Ansótegui, Felip Manyà, Je- sus Ojeda, Josep M. Salvia, and Eduard Torres. Incomplete MaxSAT approaches for combinatorial testing. Journal of Heuristics, (in press), 2022.","title":"Incomplete MaxSAT approaches for combinatorial testing","authors":"Carlos Ansótegui, Felip Manyà, Jesus Ojeda, Josep M Salvia, Eduard Torres","year":"2022","venue":"Journal of Heuristics","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Arya and Yang, 2020] Sakshi Arya and Yuhong Yang. Ran- domized allocation with nonparametric estimation for con- textual multi-armed bandits with delayed rewards. Statis- tics & Probability Letters, 164:108818, 2020.","title":"Randomized allocation with nonparametric estimation for contextual multi-armed bandits with delayed rewards","authors":"Sakshi Arya, Yuhong Yang","year":"2020","venue":"Statistics & Probability Letters","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Berg et al., 2019] Jeremias Berg, Emir Demirovic, and Pe- ter J. Stuckey. Core-boosted linear search for incomplete MaxSAT. In The Sixteenth International Conference on Integration of Constraint Programming, Artificial Intel- ligence, and Operations Research, volume 11494, pages 39-56, 2019.","title":"Core-boosted linear search for incomplete MaxSAT","authors":"Jeremias Berg, Emir Demirovic, Peter J Stuckey","year":"2019","venue":"The Sixteenth International Conference on Integration of Constraint Programming, Artificial Intelligence, and Operations Research","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Bonet et al., 2019] Blai Bonet, Guillem Francès, and Hector Geffner. Learning features and abstract actions for com- puting generalized plans. In The Thirty-Third AAAI Con- ference on Artificial Intelligence, pages 2703-2710, 2019.","title":"Learning features and abstract actions for computing generalized plans","authors":"Blai Bonet, Guillem Francès, Hector Geffner","year":"2019","venue":"The Thirty-Third AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Cai and Lei, 2020] Shaowei Cai and Zhendong Lei. Old techniques in new ways: Clause weighting, unit propaga- tion and hybridization for maximum satisfiability. Artifi- cial Intelligence, 287:103354, 2020.","title":"Old techniques in new ways: Clause weighting, unit propagation and hybridization for maximum satisfiability","authors":"Lei ; ] Shaowei Cai, Zhendong Cai, Lei","year":"2020","venue":"Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Cai et al., 2014] Shaowei Cai, Chuan Luo, John Thornton, and Kaile Su. Tailoring local search for partial MaxSAT. In The Twenty-Eighth AAAI Conference on Artificial Intel- ligence, pages 2623-2629, 2014.","title":"Tailoring local search for partial MaxSAT","authors":"Shaowei Cai, Chuan Luo, John Thornton, Kaile Su","year":"2014","venue":"The Twenty-Eighth AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Cai et al., 2017] Shaowei Cai, Chuan Luo, and Haochen Zhang. From decimation to local search and back: A new approach to MaxSAT. In The Twenty-Sixth Inter- national Joint Conference on Artificial Intelligence, pages 571-577, 2017.","title":"From decimation to local search and back: A new approach to MaxSAT","authors":"Shaowei Cai, Chuan Luo, Haochen Zhang","year":"2017","venue":"The Twenty-Sixth International Joint Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":"/proceedings/2017/80"},{"position":8,"raw":"[Cai, 2015] Shaowei Cai. Balance between complexity and quality: Local search for minimum vertex cover in massive graphs. In The Twenty-Fourth International Joint Confer- ence on Artificial Intelligence, pages 747-753, 2015.","title":"Balance between complexity and quality: Local search for minimum vertex cover in massive graphs","authors":"Shaowei Cai, Cai","year":"2015","venue":"The Twenty-Fourth International Joint Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":"/proceedings/2015/111"},{"position":9,"raw":"[Chao and Franco, 1990] Ming-Te Chao and John V. Franco. Probabilistic analysis of a generalization of the unit-clause literal selection heuristics for the k satisfiability problem. Information Sciences, 51(3):289-314, 1990.","title":"Probabilistic analysis of a generalization of the unit-clause literal selection heuristics for the k satisfiability problem","authors":"Franco Chao, Ming-Te Chao, John V Franco","year":"1990","venue":"Information Sciences","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Chvátal and Reed, 1992] Vasek Chvátal and Bruce A. Reed. Mick gets some (the odds are on his side). In The Thirty- Third Annual Symposium on Foundations of Computer Science, pages 620-627, 1992.","title":"Mick gets some (the odds are on his side)","authors":"Reed ; Chvátal, Bruce A Chvátal, Reed","year":"1992","venue":"The Thirty-Third Annual Symposium on Foundations of Computer Science","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Ciampiconi et al., 2020] Lorenzo Ciampiconi, Bishwamit- tra Ghosh, Jonathan Scarlett, and Kuldeep S. Meel. A MaxSAT-based framework for group testing. In The Thirty-Fourth AAAI Conference on Artificial Intelligence, pages 10144-10152, 2020.","title":"A MaxSAT-based framework for group testing","authors":"Lorenzo Ciampiconi, Bishwamittra Ghosh, Jonathan Scarlett, Kuldeep S Meel","year":"2020","venue":"The Thirty-Fourth AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Demirovic and Musliu, 2017] Emir Demirovic and Nysret Musliu. MaxSAT-based large neighborhood search for high school timetabling. Computers & Operations Re- search, 78:172-180, 2017.","title":"Emir Demirovic and Nysret Musliu. MaxSAT-based large neighborhood search for high school timetabling","authors":"Musliu Demirovic","year":"2017","venue":"Computers & Operations Research","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Goffinet and Ramanujan, 2016] Jack Goffinet and Raghu- ram Ramanujan. Monte-carlo tree search for the maximum satisfiability problem. In The Twenty-Second International Conference on Principles and Practice of Constraint Pro- gramming, volume 9892, pages 251-267, 2016.","title":"Monte-carlo tree search for the maximum satisfiability problem","authors":"Ramanujan Goffinet","year":"2016","venue":"The Twenty-Second International Conference on Principles and Practice of Constraint Programming","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Hu et al., 2019] Yi-Qi Hu, Yang Yu, and Jun-Da Liao. Cas- caded algorithm-selection and hyper-parameter optimiza- tion with extreme-region upper confidence bound bandit. In The Twenty-Eighth International Joint Conference on Artificial Intelligence, pages 2528-2534, 2019.","title":"Cascaded algorithm-selection and hyper-parameter optimization with extreme-region upper confidence bound bandit","authors":"Yi-Qi Hu, Yang Yu, Jun-Da Liao","year":"2019","venue":"The Twenty-Eighth International Joint Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":"/proceedings/2019/351"},{"position":15,"raw":"[Lassouaoui et al., 2019] Mourad Lassouaoui, Dalila Boughaci, and Belaid Benhamou. A multilevel synergy thompson sampling hyper-heuristic for solving Max-SAT. Intelligent Decision Technologies, 13(2):193-210, 2019.","title":"A multilevel synergy thompson sampling hyper-heuristic for solving Max-SAT","authors":"Mourad Lassouaoui, Dalila Boughaci, Belaid Benhamou","year":"2019","venue":"Intelligent Decision Technologies","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Lattimore and Szepesvári, 2020] Tor Lattimore and Csaba Szepesvári. Bandit algorithms. Cambridge University Press, 2020.","title":"Tor Lattimore and Csaba Szepesvári. Bandit algorithms","authors":"Szepesvári Lattimore","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Lei and Cai, 2018] Zhendong Lei and Shaowei Cai. Solv- ing (weighted) partial MaxSAT by dynamic local search for SAT. In The Twenty-Seventh International Joint Con- ference on Artificial Intelligence, pages 1346-1352, 2018.","title":"Solving (weighted) partial MaxSAT by dynamic local search for SAT","authors":"Cai ; ] Zhendong Lei, Shaowei Lei, Cai","year":"2018","venue":"The Twenty-Seventh International Joint Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":"/proceedings/2018/187"},{"position":18,"raw":"[Lei et al., 2021] Zhendong Lei, Shaowei Cai, Fei Geng, Dongxu Wang, Yongrong Peng, Dongdong Wan, Yip- ing Deng, and Pinyan Lu. Satlike-c: Solver description. MaxSAT Evaluation 2021, pages 19-20, 2021.","title":"Satlike-c: Solver description","authors":"Zhendong Lei, Shaowei Cai, Fei Geng, Dongxu Wang, Yongrong Peng, Dongdong Wan, Yiping Deng, Pinyan Lu","year":"2021","venue":"MaxSAT Evaluation","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Li and Manyà, 2021] Chu Min Li and Felip Manyà. MaxSAT, hard and soft constraints. In Handbook of Satis- fiability, second edition, pages 903-927. IOS Press, 2021.","title":"MaxSAT, hard and soft constraints","authors":"Min Chu, Felip Li, Manyà","year":"2021","venue":"Handbook of Satisfiability","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Li et al., 2006] Chu Min Li, Felip Manyà, and Jordi Planes. Detecting disjoint inconsistent subformulas for computing lower bounds for Max-SAT. In The Twenty-First National Conference on Artificial Intelligence, pages 86-91, 2006.","title":"Detecting disjoint inconsistent subformulas for computing lower bounds for Max-SAT","authors":"Min Chu, Felip Li, Jordi Manyà, Planes","year":"2006","venue":"The Twenty-First National Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Luo et al., 2017] Chuan Luo, Shaowei Cai, Kaile Su, and Wenxuan Huang. CCEHC: An efficient local search algo- rithm for weighted partial maximum satisfiability. Artifi- cial Intelligence, 243:26-44, 2017.","title":"CCEHC: An efficient local search algorithm for weighted partial maximum satisfiability","authors":"Chuan Luo, Shaowei Cai, Kaile Su, Wenxuan Huang","year":"2017","venue":"Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Nadel, 2019] Alexander Nadel. Anytime weighted maxsat with improved polarity selection and bit-vector optimiza- tion. In 2019 Formal Methods in Computer Aided Design, pages 193-202, 2019.","title":"Anytime weighted maxsat with improved polarity selection and bit-vector optimization","authors":"Alexander Nadel","year":"2019","venue":"Formal Methods in Computer Aided Design","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Ou et al., 2019] Mingdong Ou, Nan Li, Cheng Yang, Shenghuo Zhu, and Rong Jin. Semi-parametric sampling for stochastic bandits with many arms. In The Thirty-Third AAAI Conference on Artificial Intelligence, pages 7933- 7940, 2019.","title":"Semi-parametric sampling for stochastic bandits with many arms","authors":"Mingdong Ou, Nan Li, Cheng Yang, Shenghuo Zhu, Rong Jin","year":"2019","venue":"The Thirty-Third AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Slivkins, 2019] Aleksandrs Slivkins. Introduction to multi- armed bandits. Foundations and Trends in Machine Learn- ing, 12(1-2):1-286, 2019.","title":"Aleksandrs Slivkins. Introduction to multiarmed bandits","authors":"Slivkins","year":"2019","venue":"Foundations and Trends in Machine Learning","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}