Fermer ce champ de recherche.
Bayesian Statistics in the Big Data Era (1912)
Statistiques bayésiennes à l’ère du Big Data

Dates: 26 – 30 November 2018
Place: CIRM (Marseille Luminy, France)


Les méthodes bayésiennes sont aujourd’hui bien établies dans les domaines de la statistique et du Machine Learning et sont de plus en plus appliquées au Big Data. Toutefois, il existe encore des lacunes dans les connaissances sur la théorie, la méthodologie, le calcul et l’application des méthodes bayésiennes dans ce contexte.
Cette conférence réunira un groupe international et interdisciplinaire de chercheurs et de praticiens qui partageront leurs connaissances, leurs recherches, leurs défis et les possibilités de développer et d’utiliser les statistiques bayésiennes à l’ère des Big Data. Les résultats escomptés comprennent le transfert des connaissances, de nouveaux réseaux de collaboration, de nouvelles orientations de recherche et de nouveaux outils statistiques pour résoudre des problèmes difficiles dans le monde réel.

Bayesian methods are now firmly established in the fields of Statistics and Machine Learning and are being increasingly applied to “Big Data”. However, there are still gaps in knowledge about the theory, methodology, computation and application of Bayesian methods in this context.
This conference will bring together an international and interdisciplinary group of researchers and practitioners to share insights, research, challenges and opportunities in developing and using Bayesian statistics in the Big Data era. The anticipated outcomes include: knowledge transfer, new collaborative networks, new research directions and new statistical tools to address challenging problems in the real world.


  • Emerging Theory & Methods: Bayesian methodology for big data modelling and analysis
  • Enabling Computation: Bayesian computation for big data
  • New Insights: Applications of Bayesian analysis using big data
  • Program

The program will include oral presentations and substantial time for discussion after each presentation, as well as poster presentations with specific sessions for presentation and discussion.
The program will also include dedicated time for research discussion, collaboration and networking. Sessions on topics such as career pathways and mentoring will be scheduled for postgraduates and early career researchers.

  • Pierre Alquier (ENSAE ParisTech) – Informed Sub-Sampling MCMC: Approximate Bayesian Inference for Large Datasets  (pdf)
  • Louis Aslett (Durham University) – Privacy and Security in Pooled Bayesian Inference
  • Tamara Broderick (MIT) – Automated Scalable Bayesian Inference via Data Summarization
  • Noel Cressie (University of Wollongong)-  Inference for Spatio-Temporal Changes of Arctic Sea Ice   (pdf)  – VIDEO – 
  • Marco Cuturi (ENSAE, Paris) – Regularized Optimal Transport   (pdf)
  • David B. Dunson (Duke University) – Generalized Bayes for robust and scalable inferences from high-dimensional data   (pdf)
  • Gregor Kastner (WU Vienna) – Bayesian Inference in Many Dimensions: Examples from Macroeconomics and Finance  (pdf)
  • Ruth King (University of Edinburgh)
  • Gary Koop (University of Strathclyde) – Composite Likelihood Methods for Large Bayesian VARs with Stochastic Volatility  (pdf)
  • Antonio Lijoi (Bocconi University)  – Nonparametric priors for covariate-dependent data  (pdf)
  • Jean-Michel Marin (Université de Montpellier) – Local tree methods for classification   (pdf)
  • Antonietta Mira (Università della Svizzera italiana and University of Insubria) – Bayesian dimensionality reduction via the identifications of the data intrinsic dimensions
  • Igor Prünster (Bocconi University) – Hierarchies of discrete random probabilities   (pdf)
  • Stéphane Robin (AgroParisTech) – Shortened Bridge Sampler: Using Deterministic Approximations to Accelerate SMC for Posterior Sampling   (pdf)
  • Heejung Shim (University of Melbourne) – Bayesian multi-scale Poisson models for analyses of high-throughput sequencing data in genomics   (pdf)
  • Minh-Ngoc Tran (University of Sydney) – Bayesian Computation for Big Models Big Data
  • Darren Wilkinson (Newcastle University) – A Compositional Approach to Scalable Bayesian Computation and Probabilistic Programming   (pdf)
  • Atanu Bhattacharjee (Tata Memorial Centre) – Time-Course Data Prediction for Repeatedly Measured Gene Expression  (pdf)
  • Marta Crispino (INRIA Grenoble)  – Bayesian preference learning   (pdf)
  • Christel Faes (Hasselt University Belgium) – Accounting for residential history in disease mapping   (pdf)
  • Ethan Goan (Queensland University of Technology)
  • Logan Graham (University of Oxford) – Causality in Bayesian Modelling for Modern Machine Learning Challenges .  (pdf)
  • Clara Grazian (University of Oxford)
  • Zitong Li (University of Melbourne) – Bayesian non-parametric regression for analyzing time course quantitative genetic data  (pdf)
  • Benoit Liquet (Université de Pau et des Pays de L’Adour) – Bayesian Variable Selection Regression Of Multivariate Responses For Group Data  (pdf)
  • Jia Liu (University of Helsinki) – Bayesian model-based spatiotemporal survey design for log-Gaussian cox process  (pdf)
  • Reza Mohammadi (University of Amsterdam) –  High-dimensional Bayesian inference for Graphical Models with Application to Brain Connectivity   (pdf)
  • Ahihiko Nishimura (University of California – Los Angeles) – Computational advances in ”large n and large p » sparse Bayesian regression for binary and survival outcomes  (pdf)
  • Monica Patriche (University of Bucharest) – Equilibrium existence for Bayesian generalized games in choice form and applications
  • Pierre Pudlo (Aix-Marseille Université) – Approximate Bayesian model choice as a Machine Learning problem  (pdf)
  • Christian P. Robert (Université Paris-Dauphine) – Inference in generative models using the Wasserstein distance  (pdf)
  • Gajendra Vishwakarma (Indian Institute of Technology Dhanbad) – A Bayesian Approach for Dynamic Treatment Regimes in Presence of Competing Risk Analysis
  • Julyan Arbel (Inria Grenoble Rhône-Alpes) – Bayesian neural networks increasingly sparsify their units with depth
  • Paul-Marie Grollemund (Université de Montpellier) – Elicitation of Experts’ Knowledge for Functional Linear Regression
  • Thi Khuyen Le (Aix Marseille Université) – Connected component selection for Linear Discriminant Analysis in high dimension and applications to medical imaging
  • ​Hoang Nguyen (University Carlos III of Madrid) – Variational Inference for high dimensional structured factor copulas
  • Oluwole, K. Oyebamiji (Lancaster University) – Bayesian optimal weighting scheme for combining simulation ensemble for global climate projection
  • Maxime Rischard (Harvard University) – Unbiased estimation of log normalizing constants with applications to Bayesian cross-validation
  • Erlis Ruli ​(University of Padova) – Objective model selection with proper scoring rules and improper priors