{"type":"ScholarlyArticle","title":"Learning Bayesian Networks Under Sparsity Constraints: A Parameterized Complexity Analysis","authors":[{"name":"Niels Grüttemeier","slug":"niels-gruttemeier","given":"Niels","family":"Grüttemeier","affiliation":"Fachbereich Mathematik und Informatik , Philipps-Universität Marburg , Marburg , Germany"},{"name":"Christian Komusiewicz","slug":"christian-komusiewicz","given":"Christian","family":"Komusiewicz","affiliation":"Fachbereich Mathematik und Informatik , Philipps-Universität Marburg , Marburg , Germany"}],"abstract":"We study the problem of learning the structure of an optimal Bayesian network when additional structural constraints are posed on the network or on its moralized graph. More precisely, we consider the constraint that the moralized graph can be transformed to a graph from a sparse graph class Π by at most k vertex deletions. We show that for Π being the graphs with maximum degree 1, an optimal network can be computed in polynomial time when k is constant, extending previous work that gave an algorithm with such a running time for Π being the class of edgeless graphs [Korhonen & Parviainen, NIPS 2015]. We then show that further extensions or improvements are presumably impossible. For example, we show that when Π is the set of graphs in which each component has size at most three, then learning an optimal network is NP-hard even if k=0. Finally, we show that learning an optimal network with at most k edges in the moralized graph presumably is not fixed-parameter tractable with respect to k and that, in contrast, computing an optimal network with at most k arcs can be computed is fixed-parameter tractable in k.","doi":"10.24963/ijcai.2020/586","url":"https://ijcai.org/proceedings/2020/586","pdf":"https://ijcai.org/proceedings/2020/0586.pdf","markdown":"https://ijcai.org/proceedings/2020/586.md","bibtex":"https://ijcai.org/proceedings/2020/bibtex/586","volume":{"slug":"2020","title":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2020","url":"https://ijcai.org/proceedings/2020/","year":2020,"location":"Yokohama","dates":"January 2021 (postponed from 11-17 July 2020)","editors":"Christian Bessiere","isbn":"978-0-9992411-6-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2020"},"section":"Uncertainty in AI","track":"Main track","number":586,"pages":{"first":4245,"last":4251},"keywords":["Heuristic Search and Game Playing: Combinatorial Search and Optimisation","Uncertainty in AI: Bayesian Networks"],"videos":[{"label":"Video","url":"https://videos.ijcai20.org/24248/file_24248.m3u8","length_sec":299,"provider":"hls"}],"references":[{"position":1,"raw":"[Alon et al., 1995] Noga Alon, Raphael Yuster, and Uri Zwick. Color-coding. J. ACM, 42(4):844-856, 1995.","title":"Color-coding","authors":"Noga Alon, Raphael Yuster, Uri Zwick","year":"1995","venue":"J. ACM","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Chickering, 1995] David Maxwell Chickering. Learning Bayesian networks is NP-complete. In Proceedings of the Fifth International Conference on Artificial Intelligence and Statistics, (AISTATS'95), pages 121-130. Springer, 1995.","title":"Learning Bayesian networks is NP-complete","authors":"David Maxwell, Chickering","year":"1995","venue":"Proceedings of the Fifth International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Chow and Liu, 1968] C. K. Chow and C. N. Liu. Approxi- mating discrete probability distributions with dependence trees. IEEE Trans. Information Theory, 14(3):462-467, 1968.","title":"Approximating discrete probability distributions with dependence trees","authors":"C K Chow, C N Liu","year":"1968","venue":"IEEE Trans. Information Theory","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Cooper, 1990] Gregory F. Cooper. The computational com- plexity of probabilistic inference using Bayesian belief networks. Artif. Intell., 42(2-3):393-405, 1990.","title":"The computational complexity of probabilistic inference using Bayesian belief networks","authors":"Gregory F Cooper","year":"1990","venue":"Artif. Intell","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Cygan et al., 2015] Marek Cygan, Fedor V. Fomin, Lukasz Kowalik, Daniel Lokshtanov, Dániel Marx, Marcin Pilipczuk, Michal Pilipczuk, and Saket Saurabh. Parame- terized Algorithms. Springer, 2015.","title":"Parameterized Algorithms","authors":"Marek Cygan, V Fedor, Lukasz Fomin, Daniel Kowalik, Dániel Lokshtanov, Marcin Marx, Michal Pilipczuk, Saket Pilipczuk, Saurabh","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Darwiche, 2009] Adnan Darwiche. Modeling and Reason- ing with Bayesian Networks. Cambridge University Press, 2009.","title":"Modeling and Reasoning with Bayesian Networks","authors":"Adnan Darwiche","year":"2009","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Elidan and Gould, 2008] Gal Elidan and Stephen Gould. Learning bounded treewidth Bayesian networks. In Pro- ceedings of the Twenty-Second Annual Conference on Neural Information Processing Systems, (NIPS'08), pages 417-424.","title":"Gal Elidan and Stephen Gould. Learning bounded treewidth Bayesian networks","authors":"Gould Elidan","year":"2008","venue":"Proceedings of the Twenty-Second Annual Conference on Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"Curran Associates, Inc., 2008.","title":null,"authors":"Curran Associates, Inc","year":"2008","venue":null,"doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Gaspers et al., 2015] Serge Gaspers, Mikko Koivisto, Math- ieu Liedloff, Sebastian Ordyniak, and Stefan Szeider. On finding optimal polytrees. Theor. Comput. Sci., 592:49-58, 2015.","title":"On finding optimal polytrees","authors":"Serge Gaspers, Mikko Koivisto, Mathieu Liedloff, Sebastian Ordyniak, Stefan Szeider","year":"2015","venue":"Theor. Comput. Sci","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Korhonen and Parviainen, 2013] Janne H. Korhonen and Pekka Parviainen. Exact learning of bounded tree-width Bayesian networks. In Proceedings of the Sixteenth Inter- national Conference on Artificial Intelligence and Statis- tics, (AISTATS'13), pages 370-378. JMLR.org, 2013.","title":"Exact learning of bounded tree-width Bayesian networks","authors":"H Janne, Pekka Korhonen, Parviainen","year":"2013","venue":"Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Korhonen and Parviainen, 2015] Janne H. Korhonen and Pekka Parviainen. Tractable Bayesian network structure learning with bounded vertex cover number. In Proceed- ings of the Twenty-Eighth Annual Conference on Neural Information Processing Systems, (NIPS'15), pages 622- 630. MIT Press, 2015.","title":"Tractable Bayesian network structure learning with bounded vertex cover number","authors":"H Janne, Pekka Korhonen, Parviainen","year":"2015","venue":"Proceedings of the Twenty-Eighth Annual Conference on Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Meek, 2001] Christopher Meek. Finding a path is harder than finding a tree. J. Artif. Intell. Res., 15:383-389, 2001.","title":"Finding a path is harder than finding a tree","authors":"Christopher Meek","year":"2001","venue":"J. Artif. Intell. Res","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Naor et al., 1995] Moni Naor, Leonard J. Schulman, and Aravind Srinivasan. Splitters and near-optimal derandom- ization. In Proceedings of the Thirty-Sixth Annual Sym- posium on Foundations of Computer Science, (FOCS'95), pages 182-191. IEEE Computer Society, 1995.","title":"Splitters and near-optimal derandomization","authors":"Moni Naor, Leonard J Schulman, Aravind Srinivasan","year":"1995","venue":"Proceedings of the Thirty-Sixth Annual Symposium on Foundations of Computer Science","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Ordyniak and Szeider, 2013] Sebastian Ordyniak and Ste- fan Szeider. Parameterized complexity results for exact Bayesian network structure learning. J. Artif. Intell. Res., 46:263-302, 2013.","title":"Sebastian Ordyniak and Stefan Szeider. Parameterized complexity results for exact Bayesian network structure learning","authors":"Szeider Ordyniak","year":"2013","venue":"J. Artif. Intell. Res","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Ott and Miyano, 2003] Sascha Ott and Satoru Miyano. Finding optimal gene networks using biological con- straints. Genome Informatics, 14:124-133, 2003.","title":"Finding optimal gene networks using biological constraints","authors":"Miyano Ott, Sascha Ott, Satoru Miyano","year":"2003","venue":"Genome Informatics","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Silander and Myllymäki, 2006] Tomi Silander and Petri Myllymäki. A simple approach for finding the globally optimal Bayesian network structure. In Proceedings of the Twenty-Second Conference in Uncertainty in Artificial In- telligence (UAI'06). AUAI Press, 2006.","title":"A simple approach for finding the globally optimal Bayesian network structure","authors":"Myllymäki Silander, Tomi Silander, Petri Myllymäki","year":"2006","venue":"Proceedings of the Twenty-Second Conference in Uncertainty in Artificial Intelligence (UAI'06)","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}