{"type":"ScholarlyArticle","title":"Causal Discovery from Nonstationary/Heterogeneous Data: Skeleton Estimation and Orientation Determination","authors":[{"name":"Kun Zhang","slug":"kun-zhang","given":"Kun","family":"Zhang","affiliation":"Lingnan University , Hong Kong"},{"name":"Biwei Huang","slug":"biwei-huang","given":"Biwei","family":"Huang","affiliation":"Lingnan University , Hong Kong"},{"name":"Jiji Zhang","slug":"jiji-zhang","given":"Jiji","family":"Zhang","affiliation":"‡ Department of philosophy , Carnegie Mellon University MPI for Intelligent Systems , Tübingen , Germany"},{"name":"Clark Glymour","slug":"clark-glymour","given":"Clark","family":"Glymour","affiliation":"Lingnan University , Hong Kong"},{"name":"Bernhard Schölkopf","slug":"bernhard-scholkopf","given":"Bernhard","family":"Schölkopf","affiliation":"Lingnan University , Hong Kong"}],"abstract":"It is commonplace to encounter nonstationary or heterogeneous data, of which the underlying generating process changes over time or across data sets (the data sets may have different experimental conditions or data collection conditions). Such a distribution shift feature presents both challenges and opportunities for causal discovery. In this paper we develop a principled framework for causal discovery from such data, called Constraint-based causal Discovery from Nonstationary/heterogeneous Data (CD-NOD), which addresses two important questions. First, we propose an enhanced constraint-based procedure to detect variables whose local mechanisms change and recover the skeleton of the causal structure over observed variables. Second, we present a way to determine causal orientations by making use of independence changes in the data distribution implied by the underlying causal model, benefiting from information carried by changing distributions.  Experimental results on various synthetic and real-world data sets are presented to demonstrate the efficacy of our methods.","doi":"10.24963/ijcai.2017/187","url":"https://ijcai.org/proceedings/2017/187","pdf":"https://ijcai.org/proceedings/2017/0187.pdf","markdown":"https://ijcai.org/proceedings/2017/187.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/187","volume":{"slug":"2017","title":"Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2017","url":"https://ijcai.org/proceedings/2017/","year":2017,"location":"Melbourne, Australia","dates":"19-25 August 2017","editors":"Carles Sierra","isbn":"978-0-9992411-0-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2017"},"section":"Knowledge Representation, Reasoning, and Logic","track":"Main track","number":187,"pages":{"first":1347,"last":1353},"keywords":["Knowledge Representation, Reasoning, and Logic: Action, Change and Causality","Uncertainty in AI: Uncertainty in AI"],"videos":[],"references":[{"position":1,"raw":"[Adams and Mackay, 2007] R. P. Adams and D. J. C. Mackay. Bayesian online change point detection, 2007. Technical report, University of Cambridge, Cambridge, UK. Preprint at http://arxiv.org/abs/0710.3742v1.","title":"Bayesian online change point detection","authors":"R P Adams, D J C Mackay","year":"2007","venue":null,"doi":null,"url":"http://arxiv.org/abs/0710.3742v1","internal_url":null},{"position":2,"raw":"[Bendtsen, 2016] M. Bendtsen. Regime aware learning. In Proceedings of the Eighth International Conference on Probabilistic Graphical Models, pages 1-12, 2016.","title":"Regime aware learning","authors":"M Bendtsen","year":"2016","venue":"Proceedings of the Eighth International Conference on Probabilistic Graphical Models","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Blake and Merz, 1998] C.L. Blake and C.J. Merz. Nuclear feature extraction for breast tumor diagnosis. In UCI repository of machine learning databases, 1998.","title":"Nuclear feature extraction for breast tumor diagnosis","authors":"C L Blake, C J Merz","year":"1998","venue":"UCI repository of machine learning databases","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Calhoun et al., 2014] V. D. Calhoun, R. Miller, G. Pearlson, and T. Adal. The chronnectome: Time-varying connec- tivity networks as the next frontier in fmri data discovery. Neuron, 84:262-274, 2014.","title":"The chronnectome: Time-varying connectivity networks as the next frontier in fmri data discovery","authors":"V D Calhoun, R Miller, G Pearlson, T","year":"2014","venue":"Neuron","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Chris and Neil, 2008] M. B. Chris and B. Neil. The hip- pocampus and memory: insights from spatial processing. Nature Reviews Neuroscience, 9:182-194, 2008.","title":"The hippocampus and memory: insights from spatial processing","authors":"M B Chris, B Neil","year":"2008","venue":"Nature Reviews Neuroscience","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Chu and Glymour, 2008] T. Chu and C. Glymour. Search for additive nonlinear time series causal models. Journal of Machine Learning Research, 9:967-991, 2008.","title":"Search for additive nonlinear time series causal models","authors":"T Chu, C Glymour","year":"2008","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Havlicek et al., 2011] M. Havlicek, K.J. Friston, J. Jan, M. Brazdil, and V.D. Calhoun. Dynamic modeling of neu- ronal responses in fMRI using cubature kalman filtering. Neuroimage, 56:2109-2128, 2011.","title":"Dynamic modeling of neuronal responses in fMRI using cubature kalman filtering","authors":"M Havlicek, K J Friston, J Jan, M Brazdil, V D Calhoun","year":"2011","venue":"Neuroimage","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Hoover, 1990] K. Hoover. The logic of causal inference. Economics and Philosophy, 6:207-234, 1990.","title":"The logic of causal inference","authors":"K Hoover","year":"1990","venue":"Economics and Philosophy","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Hoyer et al., 2009] P.O. Hoyer, D. Janzing, J. Mooji, J. Pe- ters, and B. Schölkopf. Nonlinear causal discovery with additive noise models. In NIPS 21, 2009.","title":"Nonlinear causal discovery with additive noise models","authors":"P O Hoyer, D Janzing, J Mooji, J Peters, B Schölkopf","year":"2009","venue":"NIPS 21","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Huang et al., 2015] B. Huang, K. Zhang, and B. Schölkopf. Identification of time-dependent causal model: A gaussian process treatment. In Prof. IJCAI 2015, pages 3561-3568, Buenos, Argentina, 2015.","title":"Identification of time-dependent causal model: A gaussian process treatment","authors":"B Huang, K Zhang, B Schölkopf","year":"2015","venue":"Prof. IJCAI 2015","doi":null,"url":null,"internal_url":"/proceedings/2015/501"},{"position":11,"raw":"[Janzing et al., 2013] D. Janzing, D. Balduzzi, M. Grosse- Wentrup, and B. Schölkopf. Quantifying causal influences. Ann. Statist., 41:2324-2358, 2013.","title":"Quantifying causal influences","authors":"D Janzing, D Balduzzi, M Grosse-Wentrup, B Schölkopf","year":"2013","venue":"Ann. Statist","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Kummerfeld and Danks, 2013] E. Kummerfeld and D. Danks. Tracking time-varying graphical structure. In Advances in neural information processing systems 26, La Jolla, CA, 2013.","title":"Tracking time-varying graphical structure","authors":"E Kummerfeld, D Danks","year":"2013","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Madiman, 2008] M. Madiman. On the entropy of sums. In Proceedings of IEEE Information Theory Workshop (ITW'08), pages 303-307, 2008.","title":"On the entropy of sums","authors":"M Madiman","year":"2008","venue":"Proceedings of IEEE Information Theory Workshop (ITW'08)","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Pearl, 2000] J. Pearl. Causality: Models, Reasoning, and Inference. Cambridge University Press, Cambridge, 2000.","title":"Causality: Models, Reasoning, and Inference","authors":"J Pearl","year":"2000","venue":null,"doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Peters et al., 2016] J. Peters, P. Bühlmann, and N. Mein- shausen. Causal inference using invariant prediction: iden- tification and confidence intervals. Journal of the Royal Statistical Society: Series B, 2016.","title":"Causal inference using invariant prediction: identification and confidence intervals","authors":"J Peters, P Bühlmann, N Meinshausen","year":"2016","venue":"Journal of the Royal Statistical Society: Series B","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Poldrack and Laumann, 2015] R. Poldrack and T. Lau- mann. https://openfmri.org/dataset/ds000031/, 2015.","title":null,"authors":"R Poldrack, T Lau","year":null,"venue":null,"doi":null,"url":"https://openfmri.org/dataset/ds000031/,2015","internal_url":null},{"position":17,"raw":"[Shimizu et al., 2006] S. Shimizu, P.O. Hoyer, A. Hyvärinen, and A.J. Kerminen. A linear non- Gaussian acyclic model for causal discovery. Journal of Machine Learning Research, 7:2003-2030, 2006.","title":"A linear non-Gaussian acyclic model for causal discovery","authors":"S Shimizu, P O Hoyer, A Hyvärinen, A J Kerminen","year":"2006","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Song et al., 2009] L. Song, M. Kolar, and E. Xing. Time- varying dynamic Bayesian networks. In NIPS 23, 2009.","title":"Timevarying dynamic Bayesian networks","authors":"L Song, M Kolar, E Xing","year":"2009","venue":"NIPS 23","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Song et al., 2015] D. Song, MC. Hsiao, I. Opris, RE. Hamp- son, VZ. Marmarelis, GA. Gerhardt, SA. Deadwyler, and TW. Berger. Hippocampal microcircuits, functional con- nectivity, and prostheses. Recent Advances On the Modu- lar Organization of the Cortex, pages 385-405, 2015.","title":"Hippocampal microcircuits, functional connectivity, and prostheses","authors":"D Song, M C Hsiao, I Opris, R E Hampson, V Z Marmarelis, G A Gerhardt, S A Deadwyler, T W Berger","year":"2015","venue":"Recent Advances On the Modular Organization of the Cortex","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Spirtes et al., 1993] P. Spirtes, C. Glymour, and R. Scheines. Causation, Prediction, and Search. Spring-Verlag Lectures in Statistics, 1993.","title":"Causation, Prediction, and Search","authors":"P Spirtes, C Glymour, R Scheines","year":"1993","venue":"Spring-Verlag Lectures in Statistics","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Spirtes et al., 2001] P. Spirtes, C. Glymour, and R. Scheines. Causation, Prediction, and Search. MIT Press, Cambridge, MA, 2nd edition, 2001.","title":"Causation, Prediction, and Search","authors":"P Spirtes, C Glymour, R Scheines","year":"2001","venue":null,"doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Street et al., 1993] W.N. Street, W.H. Wolberg, and O.L. Mangasarian. Nuclear feature extraction for breast tu- mor diagnosis. In IS &T/SPIE's Symposium on Electronic Imaging: Science and Technology. International Society for Optics and Photonics, 1993.","title":"Nuclear feature extraction for breast tumor diagnosis","authors":"W N Street, W H Wolberg, O L Mangasarian","year":"1993","venue":"IS &T/SPIE's Symposium on Electronic Imaging: Science and Technology","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Talih and Hengartner, 2005] M. Talih and N. Hengartner. Structural learning with time-varying components: Track- ing the cross-section of financial time series. Journal of the Royal Statistical Society -Series B, 67 (3):321-341, 2005.","title":"Structural learning with time-varying components: Tracking the cross-section of financial time series","authors":"M Talih, N Hengartner","year":"2005","venue":"Journal of the Royal Statistical Society -Series B","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Tian and Pearl, 2001] J. Tian and J. Pearl. Causal discovery from changes: a bayesian approach. In Proc. UAI 2001, pages 512-521, 2001.","title":"Causal discovery from changes: a bayesian approach","authors":"J Tian, J Pearl","year":"2001","venue":"Proc. UAI 2001","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Xing et al., 2010] E. P. Xing, W. Fu, and L. Song. A state- space mixed membership blockmodel for dynamic net- work tomography. Annals of Applied Statistics, 4 (2):535- 566, 2010.","title":"A statespace mixed membership blockmodel for dynamic network tomography","authors":"E P Xing, W Fu, L Song","year":"2010","venue":"Annals of Applied Statistics","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zhang and Chan, 2006] K. Zhang and L. Chan. Extensions of ICA for causality discovery in the hong kong stock mar- ket. In Proc. 13th International Conference on Neural In- formation Processing (ICONIP 2006), 2006.","title":"Extensions of ICA for causality discovery in the hong kong stock market","authors":"K Zhang, L Chan","year":"2006","venue":"Proc. 13th International Conference on Neural Information Processing (ICONIP","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhang and Hyvärinen, 2009a] K. Zhang and A. Hyvärinen. Acyclic causality discovery with additive noise: An information-theoretical perspective. In ECML PKDD 2009, Bled, Slovenia, 2009.","title":"Acyclic causality discovery with additive noise: An information-theoretical perspective","authors":"Hyvärinen Zhang, K Zhang, A Hyvärinen","year":"2009","venue":"ECML PKDD 2009","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Zhang and Hyvärinen, 2009b] K. Zhang and A. Hyvärinen. On the identifiability of the post-nonlinear causal model. In Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, Montreal, Canada, 2009.","title":"On the identifiability of the post-nonlinear causal model","authors":"Hyvärinen Zhang, K Zhang, A Hyvärinen","year":"2009","venue":"Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Zhang et al., 2011] K. Zhang, J. Peters, D. Janzing, and B. Schölkopf. Kernel-based conditional independence test and application in causal discovery. In Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence (UAI 2011), Barcelona, Spain, 2011.","title":"Kernel-based conditional independence test and application in causal discovery","authors":"K Zhang, J Peters, D Janzing, B Schölkopf","year":"2011","venue":"Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence (UAI 2011)","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Zhang et al., 2015] K. Zhang, B. Huang, J. Zhang, B. Schölkopf, and C. Glymour. Discovery and visualiza- tion of nonstationary causal models, 2015. available at https://arxiv.org/abs/1509.08056.","title":"Discovery and visualization of nonstationary causal models","authors":"K Zhang, B Huang, J Zhang, B Schölkopf, C Glymour","year":"2015","venue":null,"doi":null,"url":"https://arxiv.org/abs/1509.08056","internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}