Research interests: probabilistic graphical models, data mining, imprecise probability, applied statistics.
My Google Scholar page.
1999: M.Sc. degree (full marks) in Environmental Engineering, Politecnico di Milano.
2002: Best Paper Prize at the International Environmental Modelling and Software Conference (Lugano, Switzerland).
2005: PhD in Information Technology (Dipartimento Elettronica ed Informazione, Politecnico di Milano).
- Thesis: "Environmental modelling via learning-from-data techniques” Advisor: Prof. G. Guariso.
- Minor thesis: "VC dimension and Structural Risk Minimization for the analysis of non-linear ecological models and time series.” Advisor: Prof. M. Gatto.
2005: visiting at Machine Learning Group - University of Bruxelles. Research topic: Model averaging and model selection for lazy learning algorithms.
2006: researcher at Cognitive Drug Research Ltd., Goring-on-Thames, Berkshire. Research topic: diagnosis of dementia through classification algorithms.
Since 2006: researcher at IDSIA.
2007: Rotary Prize (issued by Rotary Club Como, Italy) as young researcher.
Naive Credal ClassifierThe paper has been published on JMLR. The open source implementation of the naive credal classifier is the JNCC2.
Credal Model AveragingThe R implementation of credal model averaging for logistic regression is due to A. Mignatti. The algorithms are discussed in these two papers (link1, link2). The software and the marmot data set used in those papers is available here.
A Bayesian approach for comparing cross-validated algorithms on multiple data setsThe paper is published in Machine Learning, 2015 (>>preprint, >>doi). The Matlab and R software is available here.
Ongoing since 2011: Statistica Applicata, (Italian), Bach. of Management Engineering, Scuola Universitaria Professionale della Svizzera Italiana (SUPSI). Main topics: statistical inference and statistical process control.
Ongoing since 2008: Uncertain Reasoning and Data Mining (English), Master of Science in Intelligent Systems and Master in Engineering. University of Svizzera Italiana. Main topics: Bayesian networks and data mining. Co-teacher.
2000-2008: teaching assistant of Systems Analysis, Master of Science in Environmental Engineer (prof. Guariso), Polytechnic of Milan. Main topics: systems theory, simulation and identification.
IJCAI (2015, 2016): Int. Joint Conference on Artificial Intelligence
UAI (2016): Conference on Uncertainty in Artificial Intelligence
PGM (2014, 2016): European Workshop on Probabilistic Graphical Models
ECAI (2014, 2016): European Conference on Artificial Intelligence
Reviewer for Machine Learning, Int. J. of Approximate Reasoning, Int. J. of Artificial Intelligence Research and many others.
Outstanding reviewer certificate : Environmental Modelling and Software
I am co-organizer of the conference PGM 2016 , the Eighth International Conference on Probabilistic Graphical Models.
I am also co-editor of the PGM 2016 proceedings , which for the first time are published by JMLR.
A. Benavoli, G. Corani, J. Demsar, and M. Zaffalon. Time for a change: a tutorial for comparing multiple classifiers through Bayesian analysis. ArXiv e-prints 1606.04316
G. Corani, A. Benavoli, J. Demsar, F. Mangili, and M. Zaffalon. Statistical comparison of classifiers through Bayesian hierarchical modelling. ArXiv e-prints 1609.08905
A recent tutorialECML/PKDD 2016: G. Corani, A. Benavoli, J. Demsar: Comparing competing algorithms: Bayesian versus frequentist hypothesis testing, (site with slides and code).
[A.21] G. Corani and M. Scanagatta. Air pollution prediction via multi-label classification, Environmental Modelling & Software, vol. 80, pp. 259–264, 2016.
[A.20] A. Benavoli, G. Corani and F. Mangili. Should we really use post-hoc tests based on mean-ranks? Journal of Machine Learning Research, 17(5), 1-10, 2016.
[A.19] G. Corani, A. Benavoli (2015). A Bayesian approach for
comparing cross-validated algorithms on multiple data sets.
Machine Learning, vol. 100(2), pp. 285-304 (doi)
[A.17] G. Corani, A. Mignatti “Credal model averaging for classification: representing prior ignorance and expert opinions.”, International Journal of Approximate Reasoning, 2015, vol 56(B), pp. 264–277. (doi) >>preprint
[A.16] M. Zaffalon, G. Corani “Comments on “Imprecise probability models for learning multinomial distributions from data. Applications to learning credal networks” by Andrés R. Masegosa and Serafín Moral", International Journal of Approximate Reasoning, 2014, in press. (doi) >>preprint
[A.15] G. Corani, M.C. Magli, A. Giusti, L. Gianaroli, L. Gambardella, “A Bayesian Network model for predicting pregnancy after in vitro fertilization”, Computers in Biology and Medicine, 2013, 43(11), 1783–-1792. (doi). >>preprint
[A.14] L. Gianaroli, M.C. Magli, L. Gambardella, A. Giusti, C. Grugnetti and G. Corani, “Objective way to support embryo transfer: a probabilistic decision”, Human Reproduction, (2013) 28 (5): 1210-1220. (doi) >>download
[A.11] A. Giusti, P. Taddei, C. Magli, G. Corani, L. Gambardella, L. Gianaroli: "Artificial Defocus for Displaying Markers in Microscopy Z-Stacks". IEEE Transactions on Visualization and Computer Graphics, 17(12), 1757--1764, 2011. (doi).
[A.9] G.Corani, M. Zaffalon “JNCC2: The Java Implementation Of Naive Credal Classifier 2 ”, Journal of Machine Learning Research (Track on Open Source Software), 9, 2695--2698, 2008. >> download
[A.8] G.Corani, M. Zaffalon “ Learning Reliable Classifiers From Small or Incomplete Data Sets: The Naive Credal Classifier 2 ”, Journal of Machine Learning Research, 9, 581--621, 2008. >> download
[A.7] G.Corani, M. Gatto “ Structural Risk Minimization: a robust method for density-dependence detection and model selection ”, Ecography, 30(3), 400–416, 2007. >> download
[A.6] M. Bianchi, G.Corani, G. Guariso, C. Pinto “ Prediction of ungulates abundances through local linear algorithms ”, Environmental Modelling and Software, 21(10), 1508-1511, 2006 >>download
[A.5] G.Corani, M. Gatto “VC-Dimension and Structural Risk Minimization for the Analysis of Nonlinear Ecological Models”, Applied Mathematics and Computation, 176(1), 166-176, 2006 >>download
[A.3] G.Corani, G.Guariso “Coupling fuzzy modelling and neural networks for river flood prediction”, IEEE Transactions on Men, Systems and Cybernetics part C, 35(3), 382-391, 2005. >>download
[A.2] G.Corani, G.Guariso “An application of pruning in the design of neural networks for real time flood forecasting”, Neural Computing & Applications, 14, 66-77, 2005. >>download
[A.1] G.Corani “Air quality prediction in Milan: neural networks, pruned neural networks and lazy learning”, Ecological Modelling, 185, 513-529, 2005. >>download
[C.41] Scanagatta, M., Cassio de Campos, C.P., Corani G., Zaffalon, M. (2015). Learning Bayesian Networks with Thousands of Variables. Proc. Neural Information Processing Systems (NIPS 2015). Acceptance rate: 22% >>download
[C.40] A. Benavoli, G. Corani, F. Mangili and M. Zaffalon, A Bayesian nonparametric procedure for comparing algorithms. Proceedings of the 31th International Conference on Machine Learning (ICML 2015), pages 1-9, July 2015. Acceptance rate: 26% >>download
[C.39] G. Corani, A. Benavoli, F. Mangili, M. Zaffalon, Bayesian Hypothesis Testing in Machine Learning, Proceedings European Conference on Machine Learning and Knowledge Discovery in Databases, Nectar Track, (ECML PKDD 2015), pp. 199-202 >>preprint
[C.38] A. Antonucci, G. Corani, The multilabel naive credal classifier, Proceedings of Ninth International Symposium on Imprecise Probability (ISIPTA 2015), pp. 27--36 >>download
[C.37] G. Corani, C. De Campos, A Maximum Entropy Approach to Learn Bayesian Networks from Incomplete Data, Interdisciplinary Bayesian Statistics: Springer Proceedings in Mathematics & Statistics, pp. 69--82 (doi)
[C.36] C. De Campos, M. Cuccu, G. Corani and M. Zaffalon, Extended Tree Augmented Naive Classifier, Proceedings of the 7th European Workshop on Probabilistic Graphical Models (PGM 2014), pp. 176--189 >>preprint (doi)
[C.35] G. Corani, A. Antonucci, D. Maua, S. Gabaglio, Trading off speed and accuracy in multilabel classification, Proceedings of the 7th European Workshop on Probabilistic Graphical Models (PGM 2014), pp. 145--159 >>preprint (doi)
[C.34] A. E. Rizzoli, R. Rudel, A. Forster, G. Corani, F. Cellina, L.Pampuri, R. Guidi and A. Baldassari (2014) Investigating mobility styles using smartphones: advantages and limitations according to a ﬁeld study in Southern Switzerland, Proc. iEMSs 14 (7th International Congress on Environmental Modelling and Software). >>download
[C.33] A. Benavoli, F. Mangili, G. Corani, M. Zaffalon, F. Ruggeri (2014). A Bayesian Wilcoxon signed-rank test based on the Dirichlet process, Proc. ICML 14 (International Conference on Machine Learning 2014). Acceptance rate: 22% >>download >>video
[C.32] Antonucci A., Corani G., Maua D., Gabaglio S. (2013). An Ensemble of Bayesian Networks for Multilabel Classification, Proc. IJCAI 13 (23rd Int. Joint Conf. on Artificial Intelligence), 1220--1225. Acceptance rate: 28% >>download
[C.31] Corani, G., Mignatti, A. (2013). Credal model averaging of logistic regression for modeling the distribution of marmot burrows. In Cozman, F., Denoeux, T., Destercke, S., Seidenfeld, T. (Eds), Proc ISIPTA '13 (the Eighth International Symposium on Imprecise Probability: Theories and Applications). >>preprint
[C.30] A. Antonucci, G. Corani, S. Gabaglio, Active Learning by the Naive Credal Classifier, In Cano, A., Gomez-Olmedo,M., Nielsen,T. (Eds), Proc. of the 6th European Workshop on Probabilistic Graphical Models (PGM 2012), pp. 3--10. >>preprint
[C.29] G.Corani, C.Magli, A.Giusti, L. Gianaroli and L. Gambardella, A Bayesian Network model for predicting the outcome of in vitro fertilization. In Cano, A., Gomez-Olmedo,M., Nielsen,T. (Eds), Proc. of the 6th European Workshop on Probabilistic Graphical Models (PGM 2012), pp. 75--82. >>preprint
[C.28] G. Corani, A. Antonucci, and R. De Rosa, "Compression-based AODE classifiers", Proc. 20th European Conference on Artificial Intelligence (ECAI 2012), pp.264--269 . Acceptance rate: 28% >>preprint >>slides
[C.27] A. Mignatti, G. Corani, A. E. Rizzoli, "Credal Model Averaging: dealing robustly with model uncertainty on small data sets", Proc. 6th International Congress on Environmental Modelling and Software (iEMSs 2012). >>preprint
[C.26] C. Magli, G. Corani, A. Giusti, E. Castelletti, L. Gambardella and L. Gianaroli, "A prognostic model for multiple-embryo transfers", Proc. of the Annual Meeting of the European Society on Human Reproduction and Embryology (ESHRE 2012), published in Human Reproduction (Supplement: Abstract book), vol 27, suppl 2, 2012. (doi)
[C.25] A. Antonucci, M. Cattaneo, G. Corani, "Likelihood-Based Robust Classification with Bayesian Networks", Proc. of IPMU '12 , (14th Intern. Conf. on Information Processing and Management of Uncertainty in Knowledge-Based Systems), Communications in Computer and Information Science, 2012, Volume 299, Part 5, 491-500. (doi)
[C.24] A. Antonucci, M. Cattaneo, G. Corani, "Likelihood-Based Naive Credal Classifier", Proc. of ISIPTA '11 , (7th Intern. Symposium on Imprecise Probability: Theories and Applications), pages 21--30. >>download
[C.23] Marco Zaffalon, Giorgio Corani, Denis Mauá, "Utility-Based Accuracy Measures to Empirically Evaluate Credal Classifiers", Proc. of ISIPTA '11, (7th Intern. Symposium on Imprecise Probability: Theories and Applications), pages 401--410. >>download
[C.22] A. Giusti, C. Magli, G. Corani, L. Gambardella, L. Gianaroli, “Observing the 3-Dimensional Morphology of 4-Cell Embryos using Computer Analysis of Image Z-Stacks”, Proceedings of the Annual Meeting of the European Society on Human Reproduction and Embryology (ESHRE 2011), published in Human Reproduction (Supplement: Abstract book), vol.26, 2011.
[C.21] A. Giusti, G. Corani, L. Gambardella, C. Magli and L. Gianaroli, "3D Localization of Pronuclei of Human Zygotes Using Textures from Multiple Focal Planes", Proc. of the Medical Image Computing and Computer-Assisted Intervention (MICCAI 2010), Lecture Notes in Computer Science, 2010, Volume 6362/2010, pages 488-495. (acceptance rate: 32%). >>preprint
[C.20] G. Corani, A. Giusti, D. Migliore, and J. Schmidhuber, "Robust Texture Recognition Using Credal Classifiers.", Proc. of the British Machine Vision Conference 2010 (BMVC 2010), pages 78.1-78.10. (acceptance rate: 34%). >>download
[C.19] G. Corani, A. Benavoli, "Restricting
the IDM for classification: Good and evil of epsilon", Int.
Conf. on Information Processing and Management of Uncertainty
in Knowledge-Based Systems, (IPMU
2010), Part 1, pages 328--337. >>preprint
[C.18] A. Giusti, G. Corani, L.
Gambardella, C. Magli and L. Gianaroli, "Blastomere
Segmentation and 3D Morphology Measurements of Early Embryos
from Hoffman Modulation Contrast Image Stacks", IEEE
International Symposium on Biomedical Imaging (ISBI) 2010.
[C.17] A. Giusti, G. Corani, L.
Gambardella, C. Magli and L. Gianaroli “Lighting-Aware
Segmentation of Microscopy Images for In Vitro Fertilization”,
Proc. of International Symposium on Visual Computing (ISVC)
2009, Las Vegas, Springer LNCS vol. 5875/2009, 576--585.
[C.16] A. Giusti, G. Corani, L.
Gambardella, C. Magli and L. Gianaroli “Segmentation of Human
Zygotes in Hoffman Modulation Contrast Images”, Proc. of
Medical Image Understanding and Analysis (MIUA), 2009,
[C.14] G. Corani, C. De Campos, S. Yi, “A Tree Augmented Classifier Based on Extreme Imprecise Dirichlet Model”, Proc. 6th International Symposium on Imprecise Probability: Theories and Applications (ISIPTA 09), 89--98, Durham, 2009 >>download >> slides
[C.13] G.Corani, A. Rizzoli, A. Salvetti, M. Zaffalon “Reproducing human decisions in reservoir management: the case of lake Lugano”, Proc. of the 4th International ICSC Symposium on Information Technologies in Environmental Engineering, Thessaloniki, pages 252--263, 2009 >> preprint
[C.12] G.Corani, M. Zaffalon “Credal Model Averaging: an Extension of Bayesian Model Averaging to Imprecise Probabilities, Proc. ECML PKDD '08 (European Conf. on Machine Learning and Principles and Practice of Knowledge Discovery in Databases), pages 257--271 (acceptance rate: 20%) >> preprint >> slides
[C.11] G.Corani, M. Zaffalon “Naive Credal Classifier 2: an Extension of Naive Bayes for Delivering Robust Classifications”, Proc. DMIN'08 (Int. Conf. on Data Mining), Las Vegas, July 2008 >> preprint >>slides
[C.10] M. Bianchi, G.Corani, G. Guariso “PM10 forecasting with a local linear approachs”, Proc. Advanced Atmospheric Aerosol Symposium, Milano, 215-221, 2006 >>preprint
[C.9] Giorgio Corani, Chris Edgar, Isabelle Marshall, Keith Wesnes and Marco Zaffalon “Classification of dementia types from cognitive profiles data”, Proc. ECML PKDD '06 (European Conf. on Machine Learning and Principles and Practice of Knowledge Discovery in Databases), pages 470--477 (acceptance rate: 25%). >>preprint
[C.8] G.Corani, M. Gatto “An application of Structural Risk Minimization to the selection of ecological models”, 16th IFAC World Conference, Prague, July 2005
[C.7] “Fuzzy modelling of basin state and
Neural Networks for flood forecasting”, 2nd International
Environmental Modelling and Software Society Conference,
Osnabruck, June 2004. >>download
[C.6] M. Cecchetti, G. Corani, G. Guariso, “Artificial Neural Networks Prediction of PM10 in the Milan area”, 2nd International Environmental Modelling and Software Society Conference, Osnabruck, June 2004. >>download
[C.5] S. Barazzetta, G. Corani, “First results on the prediction of PM10 in Milan: the Air Sentinel project”, 9th International Conference on Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes, Garmish, May 2004.
[C.4] L.Bolognini, G. Corani, C. Faggioni,
G. Guariso, “Neural networks forecast of ozone pollution in
Milan”, 8th Int. Conference on Engineering Applications of
Neural Networks, Malaga, September 2003.
[C.3] S. Castelli, G. Corani, G. Guariso, “Structural identification of multivariate neural networks for rainfall runoff modeling”, 13th IFAC Symposium on System Identification, Rotterdam, August 2003
[C.2] 5. S. Barazzetta, G.Corani, G. Guariso, “A Neural Emission-Receptor Model for Ozone Reduction Planning”, 1st International Environmental Modelling and Software Society Conference, Lugano, 2002. This paper has been awarded with the “Best Paper Prize” of the Conference. >>download
[C.1] A.Castelletti, G.Corani, E.Weber, A. Rizzoli, R.Soncini Sessa: “A reinforcement learning approach for the operational management of a water system” IFAC Workshop on Modeling and Control in Environmental Issues. Yokohama, August 2001. >>preprint
[B.3] Corani, G., Abellán,J. , Masegosa, A., Moral,S., Zaffalon,M. “Classification”, chapter within the book Introduction to Imprecise Probabilities, p. 261-285 editors: Augustin,T., Coolen,F., de Cooman,G., Troffaes,M. (2014, in press) >> chapter preprint
[B.2] G.Corani, A. Antonucci, M. Zaffalon, “Bayesian Networks with Imprecise Probabilities: Theory and Application to Classification”, Ch. 4 of the book Data Mining: Foundations and Intelligent Paradigms , editors: D.E. Holmes, L.C. Jain, pages 49--93, 2012. >> chapter preprint
[B.1] R. Bellasio, R. Bianconi, G. Corani, G. Maffeis, M. Molari and N. Quaranta, “SINERGIE, A Decision Support System for Environmental Emergencies Management”, in "Environmental Sciences and Environmental Computing - Volume II”. E-book edited by “The Envirocomp Institute”. 2005
Invited talks and tutorials
[T.7] G. Corani, J. Demsar and A. Benavoli Comparing competing algorithms: Bayesian versus frequentist hypothesis testing, ECML/PKDD 2016 (site with slides and code).
[T.6] G. Corani, 10 years of credal classification , Sixth Workshop on Principles and Methods of Statistical Inference with Interval Probability (WPMSIIP 2013).
[T.5] A. Antonucci, G. Corani and D. Maua, Bayesian networks with imprecise probabilities: theory and applications to knowledge-based systems and classification, IJCAI '13 (23rd Int. Joint Conf. on Artificial Intelligence, 2013. [T.4] A. Antonucci, C.P. De Campos and G. Corani, “Classification with Imprecise Probabilities”, 4th SIPTA Summer School , Durham, September 2010. >> All slides of the school.
[T.3] A. Antonucci, G. Corani, “ “Bayesian Networks with Imprecise Probabilities: Theory and Applications to Knowledge-based Systems and Classification”,”: “AAAI-10: Twenty-Fourth Conference on Artificial Intelligence”, Atlanta, July 2010.
[T.2] G. Corani, “Statistical approaches for air pollution prediction”, within the workshop: “Méthodes Statistiques et Pollution”, Insa de Rouen, June 2008 >> slides >> handouts
[T.1] G. Corani, “Naive Credal Classifier”, within the “SIPTA (Society for Imprecise Probability: Theories and Applications) School 2008",, (15 mins presentation, within the more general tutorial about credal networks), Montpellier, July 2008 >> slides