liu.seSearch for publications in DiVA
Change search
Link to record
Permanent link

Direct link
BETA
Peña, Jose M.
Alternative names
Publications (10 of 56) Show all publications
Peña, J. M. (2018). Identification of Strong Edges in AMP Chain Graphs. In: Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence (UAI 2018): . Paper presented at the 34th Conference on Uncertainty in Artificial Intelligence (UAI 2018), Monterey, California, USA, August 6-10, 2018.
Open this publication in new window or tab >>Identification of Strong Edges in AMP Chain Graphs
2018 (English)In: Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence (UAI 2018), 2018Conference paper, Published paper (Refereed)
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-159303 (URN)
Conference
the 34th Conference on Uncertainty in Artificial Intelligence (UAI 2018), Monterey, California, USA, August 6-10, 2018
Available from: 2019-08-06 Created: 2019-08-06 Last updated: 2019-08-16Bibliographically approved
Peña, J. M. (2018). Reasoning with Alternative Acyclic Directed Mixed Graphs. Behaviormetrika, 45(2), 389-422
Open this publication in new window or tab >>Reasoning with Alternative Acyclic Directed Mixed Graphs
2018 (English)In: Behaviormetrika, ISSN 0385-7417, E-ISSN 1349-6964, Vol. 45, no 2, p. 389-422Article in journal (Refereed) Published
Abstract [en]

Acyclic directed mixed graphs (ADMGs) are the graphs used by Pearl (Causality: models, reasoning, and inference. Cambridge University Press, Cambridge, 2009) for causal effect identification. Recently, alternative acyclic directed mixed graphs (aADMGs) have been proposed by Peña (Proceedings of the 32nd conference on uncertainty in artificial intelligence, 577–586, 2016) for causal effect identification in domains with additive noise. Since the ADMG and the aADMG of the domain at hand may encode different model assumptions, it may be that the causal effect of interest is identifiable in one but not in the other. Causal effect identification in ADMGs is well understood. In this paper, we introduce a sound algorithm for identifying arbitrary causal effects from aADMGs. We show that the algorithm follows from a calculus similar to Pearl’s do-calculus. Then, we turn our attention to Andersson–Madigan–Perlman chain graphs, which are a subclass of aADMGs, and propose a factorization for the positive discrete probability distributions that are Markovian with respect to these chain graphs. We also develop an algorithm to perform maximum likelihood estimation of the factors in the factorization.

Place, publisher, year, edition, pages
Tokyo, Japan: Nihon Kodo Keiryo Gakkai, 2018
Keywords
Causality, Causal effect identification, Acyclic directed mixed graphs, Factorization, Maximum likelihood estimation
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-159350 (URN)10.1007/s41237-018-0051-2 (DOI)
Available from: 2019-08-08 Created: 2019-08-08 Last updated: 2019-08-15Bibliographically approved
Peña, J. M. (2018). Unifying DAGs and UGs. In: Proceedings of the 9th International Conference on Probabilistic Graphical Models (PGM 2018) - Proceedings of Machine Learning Research 72: . Paper presented at the 9th International Conference on Probabilistic Graphical Models (PGM 2018), Prague, Czech Republic, September 11 - 14, 2018.
Open this publication in new window or tab >>Unifying DAGs and UGs
2018 (English)In: Proceedings of the 9th International Conference on Probabilistic Graphical Models (PGM 2018) - Proceedings of Machine Learning Research 72, 2018Conference paper, Published paper (Refereed)
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-159302 (URN)
Conference
the 9th International Conference on Probabilistic Graphical Models (PGM 2018), Prague, Czech Republic, September 11 - 14, 2018
Available from: 2019-08-06 Created: 2019-08-06 Last updated: 2019-08-16Bibliographically approved
Peña, J. M. (2017). Causal Effect Identification in Alternative Acyclic Directed Mixed Graphs. In: Proceedings of the 3rd Workshop on Advanced Methodologies for Bayesian Networks (AMBN 2017) - Proceedings of Machine Learning Research 73, 21-32: . Paper presented at the 3rd Workshop on Advanced Methodologies for Bayesian Networks (AMBN 2017), Kyoto, Japan, 20-22 September 2017.
Open this publication in new window or tab >>Causal Effect Identification in Alternative Acyclic Directed Mixed Graphs
2017 (English)In: Proceedings of the 3rd Workshop on Advanced Methodologies for Bayesian Networks (AMBN 2017) - Proceedings of Machine Learning Research 73, 21-32, 2017Conference paper, Published paper (Refereed)
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-159353 (URN)
Conference
the 3rd Workshop on Advanced Methodologies for Bayesian Networks (AMBN 2017), Kyoto, Japan, 20-22 September 2017
Available from: 2019-08-08 Created: 2019-08-08 Last updated: 2019-08-16Bibliographically approved
Peña, J. M. (2017). Learning Causal AMP Chain Graphs. In: Proceedings of the 3rd Workshop on Advanced Methodologies for Bayesian Networks (AMBN 2017) - Proceedings of Machine Learning Research 73, 33-44.: . Paper presented at the 3rd Workshop on Advanced Methodologies for Bayesian Networks (AMBN 2017), Kyoto, Japan, 20-22 September 2017.
Open this publication in new window or tab >>Learning Causal AMP Chain Graphs
2017 (English)In: Proceedings of the 3rd Workshop on Advanced Methodologies for Bayesian Networks (AMBN 2017) - Proceedings of Machine Learning Research 73, 33-44., 2017Conference paper, Published paper (Refereed)
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-159354 (URN)
Conference
the 3rd Workshop on Advanced Methodologies for Bayesian Networks (AMBN 2017), Kyoto, Japan, 20-22 September 2017
Available from: 2019-08-08 Created: 2019-08-08 Last updated: 2019-08-16Bibliographically approved
Bendtsen, M. & Peña, J. M. (2017). Modelling regimes with Bayesian network mixtures. In: Niklas Lavesson (Ed.), Proceedings of the 30th Annual Workshop of the Swedish Artificial Intelligence Society SAIS 2017, May 15–16, 2017, Karlskrona, Sweden: . Paper presented at The 30th Annual Workshop of the Swedish Artificial Intelligence Society SAIS 2017, May 15–16, 2017, Karlskrona, Sweden (pp. 20-29). Linköping: Linköping University Electronic Press, 137, Article ID 002.
Open this publication in new window or tab >>Modelling regimes with Bayesian network mixtures
2017 (English)In: Proceedings of the 30th Annual Workshop of the Swedish Artificial Intelligence Society SAIS 2017, May 15–16, 2017, Karlskrona, Sweden / [ed] Niklas Lavesson, Linköping: Linköping University Electronic Press, 2017, Vol. 137, p. 20-29, article id 002Conference paper, Published paper (Refereed)
Abstract [en]

Bayesian networks (BNs) are advantageous when representing single independence models, however they do not allow us to model changes among the relationships of the random variables over time. Due to such regime changes, it may be necessary to use different BNs at different times in order to have an appropriate model over the random variables. In this paper we propose two extensions to the traditional hidden Markov model, allowing us to represent both the different regimes using different BNs, and potential driving forces behind the regime changes, by modelling potential dependence between state transitions and some observable variables. We show how expectation maximisation can be used to learn the parameters of the proposed model, and run both synthetic and real-world experiments to show the model’s potential.

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2017
Series
Linköping Electronic Conference Proceedings, ISSN 1650-3686, E-ISSN 1650-3740 ; 137
Keywords
Bayesian networks, hidden Markov models, regimes, algorithmic trading
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-137664 (URN)9789176854969 (ISBN)
Conference
The 30th Annual Workshop of the Swedish Artificial Intelligence Society SAIS 2017, May 15–16, 2017, Karlskrona, Sweden
Available from: 2017-05-24 Created: 2017-05-24 Last updated: 2019-07-04Bibliographically approved
Peña, J. M. (2016). Alternative Markov and Causal Properties for Acyclic Directed Mixed Graphs. In: Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence (UAI 2016): . Paper presented at The 32nd Conference on Uncertainty in Artificial Intelligence (UAI 2016), New York City, NY, USA, June 25-29, 2016.
Open this publication in new window or tab >>Alternative Markov and Causal Properties for Acyclic Directed Mixed Graphs
2016 (English)In: Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence (UAI 2016), 2016Conference paper, Published paper (Refereed)
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-141894 (URN)978-0-9966431-1-5 (ISBN)
Conference
The 32nd Conference on Uncertainty in Artificial Intelligence (UAI 2016), New York City, NY, USA, June 25-29, 2016
Available from: 2017-10-12 Created: 2017-10-12 Last updated: 2018-01-13Bibliographically approved
Bendtsen, M. & Peña, J. M. (2016). Gated Bayesian Networks for Algorithmic Trading. International Journal of Approximate Reasoning, 69, 58-80
Open this publication in new window or tab >>Gated Bayesian Networks for Algorithmic Trading
2016 (English)In: International Journal of Approximate Reasoning, ISSN 0888-613X, E-ISSN 1873-4731, Vol. 69, p. 58-80Article in journal (Refereed) Published
Abstract [en]

Gated Bayesian networks (GBNs) are a recently introduced extension of Bayesian networks that aims to model dynamical systems consisting of several distinct phases. In this paper, we present an algorithm for semi-automatic learning of GBNs. We use the algorithm to learn GBNs that output buy and sell decisions for use in algorithmic trading systems. We show how using the learnt GBNs can substantially lower risks towards invested capital, while at the same time generating similar or better rewards, compared to the benchmark investment strategy buy-and-hold.

Place, publisher, year, edition, pages
Elsevier: , 2016
Keywords
Probabilistic graphical models, Bayesian networks, algorithmic
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-124066 (URN)10.1016/j.ijar.2015.11.002 (DOI)000368957000004 ()
Note

Funding agencies: Center for Industrial Information Technology, Linkoping University (CENIIT) [09.01]; Swedish Research Council [2010-4808]

Available from: 2016-01-19 Created: 2016-01-19 Last updated: 2018-01-10
Peña, J. M. (2016). Learning Acyclic Directed Mixed Graphs from Observations and Interventions. In: Proceedings of the 8th International Conference on Probabilistic Graphical Models (PGM 2016) - JMLR: Workshop and Conference Proceedings 52: . Paper presented at The 8th International Conference on Probabilistic Graphical Models (PGM 2016), Lugano, Switzerland, September 6-9, 2016.
Open this publication in new window or tab >>Learning Acyclic Directed Mixed Graphs from Observations and Interventions
2016 (English)In: Proceedings of the 8th International Conference on Probabilistic Graphical Models (PGM 2016) - JMLR: Workshop and Conference Proceedings 52, 2016Conference paper, Published paper (Refereed)
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-141895 (URN)
Conference
The 8th International Conference on Probabilistic Graphical Models (PGM 2016), Lugano, Switzerland, September 6-9, 2016
Available from: 2017-10-12 Created: 2017-10-12 Last updated: 2018-01-13Bibliographically approved
Pena, J. M. & Gomez-Olmedo, M. (2016). Learning marginal AMP chain graphs under faithfulness revisited. International Journal of Approximate Reasoning, 68, 108-126
Open this publication in new window or tab >>Learning marginal AMP chain graphs under faithfulness revisited
2016 (English)In: International Journal of Approximate Reasoning, ISSN 0888-613X, E-ISSN 1873-4731, Vol. 68, p. 108-126Article in journal (Refereed) Published
Abstract [en]

Marginal AMP chain graphs are a recently introduced family of models that is based on graphs that may have undirected, directed and bidirected edges. They unify and generalize the AMP and the multivariate regression interpretations of chain graphs. In this paper, we present a constraint based algorithm for learning a marginal AMP chain graph from a probability distribution which is faithful to it. We show that the marginal AMP chain graph returned by our algorithm is a distinguished member of its Markov equivalence class. We also show that our algorithm performs well in practice. Finally, we show that the extension of Meeks conjecture to marginal AMP chain graphs does not hold, which compromises the development of efficient and correct score+search learning algorithms under assumptions weaker than faithfulness. (C) 2015 Elsevier Inc. All rights reserved.

Place, publisher, year, edition, pages
ELSEVIER SCIENCE INC, 2016
Keywords
Chain graphs; AMP chain graphs; MVR chain graphs; Structure learning
National Category
Computer and Information Sciences
Identifiers
urn:nbn:se:liu:diva-124106 (URN)10.1016/j.ijar.2015.09.004 (DOI)000366774200009 ()
Note

Funding Agencies|Center for Industrial Information Technology [09.01]; Swedish Research Council [2010-4808]; Spanish Ministry of Economy and Competitiveness [TIN2013-46638-C3-2-P]; European Regional Development Fund (FEDER)

Available from: 2016-01-25 Created: 2016-01-19 Last updated: 2018-01-10
Organisations

Search in DiVA

Show all publications