James Cussens

Picture: On Craig-y-dorth with Sugar Loaf (Mynydd Pen-y-fâl) in the background.

[Research] [PhD supervision] [Projects] [Software] [Teaching] [Professional Activities] [Pastoral and admin roles]
[Personal history] [Contact information]

There appears to me to be a difficulty in this conclusion: that happenings which depend upon an infinite number of cases cannot be determined by a finite number of experiments; indeed nature has her own habits, born from the return of causes, but only 'in general'. And so, who will say whether a subsequent experiment will not stray somewhat from the rule of all the preceding experiments, because of the very mutabilities of things? [Letter from Leibniz to Bernoulli, 3 December 1703. Quoted in: Cussens, Probability and Statistics in Antognazza (ed.) The Oxford Handbook of Leibniz, OUP, 2018.]

Research

GOBNILP software for Bayesian network structure learning

Google scholar profile

University of Bristol research profile

Recent papers

• David Beer, Alison Wallace, Roger Burrows, Alexandra Ciocanel and James Cussens. Valuing the manual: the demarcation of embodied practices within algorithmic decision-making processes. Social & Cultural Geography, 1–19, 2024.
• Alexandra Ciocănel, Alison Wallace, David Beer, James Cussens and Roger Burrows. Open Banking and data reassurance: the case of tenant referencing in the UK. Information, Communication and Society, 2024.
• David Beer, Alison Wallace, Alexandra Ciocanel, Roger Burrows and James Cussens. Automation hesitancy: confidence deficits, established limits and notional horizons in the application of algorithms within the private rental sector in the UK. Information, Communication & Society, 2023.
• James Cussens. Branch-Price-and-Cut for Causal Discovery. Proc. 2nd Conference on Causal Learning and Reasoning (CLeaR 2023), PMLR, 2023, (Poster).
• Milan Studený, James Cussens and Václav Kratochvíl. The dual polyhedron to the chordal graph polytope and the rebuttal of the chordal graph conjecture. International Journal of Approximate Reasoning, Volume 138, November 2021, Pages 188-203, November 2021.
• James Cussens. GOBNILP: Learning Bayesian network structure with integer programming (software demo). Proceedings of the The 10th International Conference on Probabilistic Graphical Models (PGM 2020)
• Milan Studený, James Cussens and Václav Kratochvíl. Dual Formulation of the Chordal Graph Conjecture. Proceedings of the The 10th International Conference on Probabilistic Graphical Models (PGM 2020)
• Teny Handhayani and James Cussens. Kernel-based Approach for Learning Causal Graphs from Mixed Data. Proceedings of the The 10th International Conference on Probabilistic Graphical Models (PGM 2020)
• Charupriya Sharma, Zhenyu Liao, James Cussens and Peter van Beek. A Score-and-Search Approach to Learning Bayesian Networks with Noisy-OR Relations . Proceedings of the The 10th International Conference on Probabilistic Graphical Models (PGM 2020)
• Kocacoban, D., Cussens, J. Fast Online Learning in the Presence of Latent Variables. Digitale Welt 4, 37–42 (2020), 2020
• Alvaro H. C. Correia, James Cussens and Cassio de Campos. On Pruning for Score-Based Bayesian Network Structure Learning. Proc. AISTATS 2020 and arXiv 1905.09943, May 2019.
• D. Kocacoban and J. Cussens, "Online Causal Structure Learning in the Presence of Latent Variables," 2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA), Boca Raton, FL, USA, 2019, pp. 392-395, doi: 10.1109/ICMLA.2019.00073.
• Christopher M. Hatton, Lewis W. Paton, Dean McMillan, James Cussens, Simon Gilbody and Paul A. Tiffin. Predicting persistent depressive symptoms in older adults: a machine learning approach to personalised mental healthcare. Journal of Affective Disorders, 2018.
• Zhenyu A. Liao, Charupriya Sharma, James Cussens and Peter van Beek. Finding All Bayesian Network Structures within a Factor of Optimal. Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19), 2019.
• James Cussens. Markov Random Field MAP as Set Partitioning. Proceedings of the Ninth International Conference on Probabilistic Graphical Models (PGM'18), PMLR 72:85-96, 2018.
• James Cussens. Finding Minimal Cost Herbrand Models with Branch-Cut-and-Price. arXiv:1808.04758, August 2018

Recent talks

• Branch-price-and-cut for Bayesian network structure learning, SCIP 20th Anniversary Workshop School, ZIB, Berlin, 4 November 2022.
• Algorithms for learning Bayesian networks, Interactive AI CDT Summer School (BIAS 22), University of Bristol, 6 September 2022.
• Bayesian network structure learning using integer programming, ISL/Interactive AI CDT Research seminar, University of Bristol, 10 November 2020 (online). Video has limited access, here are the slides
• Markov Random Field MAP as Set Partitioning, PGM'18, Prague, 13 September 2018.
• Towards the Holy Grail in Machine Learning, CP'18 (Invited talk), Lille, 29 August 2018.

PhD supervision

I am seeking suitably qualified PhD students and have a page with suggested PhD topics. I am seeking suitably qualified PhD students and If you are interested in probabilistic graphical models, integer programming applications in machine learning or statistical relational learning then contact me.

Former students

• Teny Handhayani - A Kernel-based Approach for Learning Causal Graphs From Mixed Data Containing Missing Values
• Durdane Kocacoban - Online Causal Structure Learning in the Presence of Latent Variables
• Mark Balmer (MSc by Research)
• Garo Panikian - Statistical inference of dynamical systems with application to modelling fish populations
• Eman Aljohani - Informative priors for learning graphical models
• Waleed Alsanie - Learning PRISM programs
• Joanne Powell - PrediCtoR: Predicting the Recovery of Ancient DNA and Ancient Proteins (with Matthew Collins, Archaeology)
• Adel Aloraini - Extending the graphical represetation of KEGG pathways for a better understanding of prostate cancer using machine learning
• Barnaby Fisher - Inductive Logic Programming and Mercury (MSc by Research)
• Heather Maclaren - Inductive Logic Programming for Software Agents: Algorithms and Implementations

Software

Use the following links to get the C/SCIP and Python/Gurobi versions of GOBNILP, respectively:

• GOBNILP software for exact Bayesian network learning (C / SCIP version)
• GOBNILP software for exact Bayesian network learning (Python / Gurobi version )

• Publications using GOBNILP
• SCIP
• pygobnilp API documentation
• pygobnilp manual
• pygobnilp Jupyter notebooks
• Benchpress, a Snakemake workflow for running and evaluating a number of Bayesian network structure learning algorithms, including GOBNILP.
• Here you can find a zip archive of the files associated with the paper Bayesian network learning with cutting planes Proc. UAI 2011. These are the files mentioned in Section 5.1 of that paper which are no longer available via the URL given in the paper.

Projects

• Code encounters: algorithmic risk-profiling in housing. Jan 2022 - July 2024. Funded by the Nuffield Foundation.
• Tailoring health policies to improve outcomes using machine learning, causal inference and operations research methods 01/07/20-30/06/23. Funded by MRC.
• Imprecision and importance: probabilistic graphical models in toxicology. 01/07/13-01/12/15. Funded by NC3RS.
• A graphical model approach to pedigree construction using constrained optimisation 03/10/11-01/02/15. Funded by the MRC.
• Stochastic Logic Programs for MCMC (with Nicos Angelopoulos) 01/09/03--31/08/05. Funded by EPSRC's MathFIT programme.
• APRIL II: Applications of Probabilistic Inductive Logic Programming 02/01/04--02/01/07. Funded by the EU IST programme.

Teaching

Currently, at Bristol:

• Machine Learning
• Foundations of AI

Professional Activities

General chair

• UAI23 (with Kun Zhang)

Programme chair

• UAI22 (with Kun Zhang)
• ILP16 (with Alessandra Russo)
• ILP00 (with Alan Frisch)
• LLL99 LLL05 (with Claire Nédellec)

Invited speaker

• Bilbao Data Science Workshop, Bilbao, November 2019
• Graphical Models: Conditional Independence and Algebraic Structures, Munich, October 2019
• CP 2018, Lille, 27-31 August, 2018
• Workshop on Learning with Structured Data and Natural Language, Toulouse, 9-11 December, 2015.
• Joint Workshop on Limit Theorems and Algebraic Statistics, Prague Stochastics 2014, August 25-29, 2014
• ICLP workshop on Probabilistic logic programming, 17 July 2014
• ILP-MLG-SRL 09
• UKKDD-2007
• AC05
• ILP04

Area chair/Senior PC

• AAAI 2025,
• AAAI 2024, IJCAI 2024, UAI 2024
• IJCAI 2023, AAAI 2023
• IJCAI-ECAI 2022
• IJCAI-21, UAI 2021, ECML/PKDD 21
• ECAI 2020
• IJCAI-19
• IJCAI-ECAI-18
• IJCAI-17
• IJCAI-15 (Main track), IJCAI-15 (Machine Learning track)
• IJCAI-13
• ECAI-12
• IJCAI-11, AAAI-11
• ECML/PKDD-08
• ECML/PKDD-07
• ECML/PKDD-06
• ICML05

Co-organiser

• Session on Combinatorial optimization methods for graphical model selection at CMStatistics 2015, London, 12-14 December 2015
• ICMS workshop on Statistical and computational methods for relatedness and relationship inference from genetic marker data, Sept 2014, with Nuala Sheehan
• IJCAI 2013 Tutorial on Optimal Algorithms for Learning Bayesian Networks with Brandon Malone and Changhe Yuan
• ICML 2012 Workshop on Statistical Relational Learning (SRL-12) together with Kristian Kersting, Sriraam Natarajan, and David Poole.

PC member / Reviewer

• PLP 2024, PGM 2024, IJCLR 2024
• IJCLR 2023
• PGM 2022, ILP 2022
• NeurIPS 2021 (Outstanding Reviewer Award), ICML 2021, AISTATS 2021, ILP2020-21@IJCLR PLP 2021
• NeurIPS 2020, AAAI-20, AISTATS 2020, UAI 2020, PGM 2020, ILP 2020
• AAAI-19, AISTATS 2019, ICML-19, ILP 2019
• NIPS18 (Outstanding Reviewer Award), ICML-18, AAAI-18, AISTATS 2018, ICLR 2018, PGM 2018, UAI 2018, ILP 2018
• NIPS17, AAAI-17, AISTATS 2017, ICML 2017, ECML/PKDD 2017, UAI 2017, ILP 2017
• NIPS16, AAAI-16 , KDD 2016 , UAI 2016 , IJCAI-16 , ECML/PKDD 2016 , ECAI 2016 , StarAI 2016 , PGM 2016 , PLP 2016
• NIPS15, ECML/PKDD 2015,UAI 2015, AAAI-15, ILP 2015, PLP 2015
• NIPS14, ICML 2014, UAI 2014, ILP 2014, ECML/PKDD 2014, AAAI-14, KR 2014, ECAI'14, BUDA 2014,
• NIPS13, ICML 2013, UAI 2013, ILP 2013, ECML/PKDD 2013, EMNLP 2013, NAACL-HLT 2013, LML workshop at ECML/PKDD 2013
• NIPS12, ICML 2012, UAI 2012, ILP 2012, ECML/PKDD 2012, AAAI-12, KR 2012 StaRAI-12, CoCoMile 2012, ACL 2012, Cognitive 2012,
• ICML 2011, UAI 2011, ILP 2011, ECML/PKDD 2011,
• ILP 2010, AAAI-10, ECAI-2010, ECML/PKDD 2010, SBIA 2010
• NIPS09, EACL09, ICML 09, ILP-09, SRL-09, Terminologie et intelligence artificielle (TIA - 2009), IJCAI-09, AISTATS 09, CoNLL 09, NAACL-HLT 09, EACL Cognitive 2009, NAACL-2009 Workshop on Unsupervised and Minimally Supervised Learning of Lexical Semantics
• NIPS08, ICML 08, ILP 08, ECAI 08, SBIA 08, CoNLL 08,
• ICML 07, UAI07, ILP07, ACL-2007 Workshop on Cognitive Aspects of Computational Language Acquisition, TIA'07
• EACL06, UAI06, ILP06, AAAI-06, SRL06, CoNLL06
• IJCAI05, ICML05, UAI05, ILP05, ECML/PKDD05, LLLL, CoNLL05, TIA05
• NIPS04, ICML04, UAI04, ECML04, CIFT04, SRL04, CoNLL04, Psycho-computational models ...
• ICML03, UAI03, ILP03, CoNLL03, Acquisition, apprentissage et ..., SRL2003, ECML03
• ICML02, UAI02, ILP02, CIFT02, CoNLL02
• ILP01, ECML01 , CoNLL01, LLL01
• ILP00, CoNLL00, LLL00
• ILP99 , LLL99
• ILP98

Miscellaneous

• Participant at the Simons Institute programme on Causality at UC Berkeley.
• Attendee at the 2014 Dagstuhl seminar on Constraints, Optimization and Data
• Speaker at the specialists' workshop of the ENBIS-SFdS 2014 Spring Meeting on Graphical causality models: Trees, Bayesian Networks and Big Data
• Attendee at the 2011 Dagstuhl seminar on Constraint Programming meets Machine Learning and Data Mining
• Attendee at the 2007 Dagstuhl seminar on Probabilistic, Logical and Relational Learning - A Further Synthesis
• Attendee at the 2005 Dagstuhl seminar on Probabilistic, Logical and Relational Learning - Towards a Synthesis

Pastoral and admin roles

• Lead Senior Tutor, School of Computer Science

Personal history

Contact information