Research

Statistical methods for complex data, and their translation into machine learning and deep learning — from biomedical omics to coastal and environmental science.

Topics

Since 1999, my research has developed statistical methods for complex data, including model selection, survival analysis, and dimension reduction, with applications to clinical, epidemiological, and environmental studies, particularly in coastal hazard phenomena. Building on this foundation, I now specialize in machine learning and deep learning, adapting statistical rigour to develop robust AI methods. My work spans medical research (analyzing omics and high-dimensional biomedical data) to environmental science (modeling coastal risks and extreme events), ensuring interpretable and reliable results across domains.

Areas of expertise

Partial least squares. Flexible PLS models exploiting penalties that perform variable selection in complex situations such as nested group structure and longitudinal designs, now scalable to big data. Six publications and four R packages.
Sliced inverse regression. More than ten publications on SIR methods, including BIG-SIR, a divide-and-conquer estimator for massive datasets with established theoretical properties.
Bayesian variable selection. Sparse Bayesian approaches for high-dimensional data using evolutionary stochastic search and hierarchical spike-and-slab priors, including the first Bayesian meta-analysis models incorporating pathway-level group structure for genomics. Awarded the Lindley Prize in 2018.
Features selection. Fast statistical tool used to find the best group of predictor variables for regression models
Deep learning. Mathematical foundations and applications, from hyperspectral imaging in brain tumour surgery to super-resolution models for coastal sea states.

Funded projects

2015-2016. “EOLE” project Funded by UPPA as a BQR (“Bonus Qualité Recherche”) EOLE aims to introduce particularly innovative action that involves the synthesis of knowledge and analysis of metadata on the environment and fishery resources of the “Golf de Gascogne”. Co-supervision with Noelle Bru (LMAP), Gilles Morandeau and Nathalie Caill-Milly (Ifremer) of a research internship student (6 months).
2015-2017 “Mechanomics” project “Investigating molecular mechanisms involved in smoking-induced lung cancer”. This project granted by the Cancer Research UK calls will extend novel mathematical modelling approaches to integrate exposure history, genetic variants, and identify the pathological stage(s) at which they exert their effects, and if/when they contribute to the dynamics of disease progression. Our international project involves partners from Europe (UK, The Netherlands, France) and from the USA. .
2016-2021. “MICROPOLIT” project This projet supported by all teams from MIRA aims to “revisit” the concept of quality of the coastal environment in a multi-disciplinary approach and spatio-temporal to construct indicators, allowing managers a more global view, more ecosystem and thereby foster a more meaningful management. It aims to assess the past and current quality of the environment, but also to predict future changes and to propose possible remedies measures. I was involved in several actions of this project and the coordinator of one action (“ Modelisation and Simulation”)
2016-2019 “ASMAH” project ASMAH “Advanced Statistical Methods for the analysis of High-dimensional data” aims to boost and enhance the current toolkit for integrative analysis of massive datasets, especially for longitudinal design. Using high resolution/ high quality real data this project will generate innovative methodologies that overcome the big data challenge in a real-life context and thus help the full exploration of existing and yet under exploited data.
2017-2019 “CRA DSIT” project Partner Investigator of project entitled “Innovation for Data and Analytics Workflows” which aims to analyse anomaly detection and correction in water quality sensors parameters.
2019-2021 “CCGIP” project Principal Investigator of project entitled “Cross Cancer Genomic Investigation of Pleiotropy Effect at Pathway and GxE interactions levels: application to breath and thyroid cancers” which has been granted by “ La Ligue contre le Cancer”. This project involves 3 teams: Inserm U1018 (Epidemiology of cancers: genes and environment), LMAP and Imperial College London (Department of Mathematics). The MRC Biostatistics Unit in Cambridge is also a partner of this project. I was the coordinator of the consortium.
2018-2021 “BIGCEE” project Principal Investigator of project entitled “Big model and Big data in Computational Ecology and Environmental Sciences” granted by the “E2S-UPPA 2018 New challenges”. This project puts forward innovative and novel methodology to develop tools that will overcome the big data challenge in Ecology and Environmental Sciences and thus reveal nature's secrets, currently hidden in big data.
2020-2022 “Linkage” project Partner Investigator of project entitled “Revolutionising water-quality monitoring in the information age” granted by the “Linkage ARC grant 2020”. This project aims to develop novel statistical methods to detect anomalies in the data generated from these in-situ sensors with computationally efficient modelling on river networks through space and time.
2019-2022 “GOLD project” Partner Investigator of the project entitled “GenOmic variability in heaLth & Disease” granted by INSERM regarding the CROSS-CUTTING PROGRAM 2018 call. The aim of the GOLD Cross-Cutting Program is to provide frameworks and tools to extract relevant information from these data to understand links between genomic variability and disease onset and progression.
2020-2024 “AMLAP” project Leader of the project entitled “Advanced Machine Learning Algorithms for leveraging Pleiotropy effect (AMLAP)” which has been granted and elected to the 2020 call for proposals from “Interdisciplinary approaches in oncogenic processes and therapeutic perspectives: Contributions of mathematics and informatics to oncology” (funded 520,000 euros). The novel methodology that we aim to develop will enable scientists to explore and analyse massive data sets from the "dark genome", majority of genes in the genome have minimal knowledge. Two Postdocs have been funded for 3 years. I was leading the project which involves a team from INSERM Paris.
2022–2025 Chief Investigator ($405,000) ARC Discovery Project “High Predictive Performance Models via Semi-Parametric Survival Regression”. This project will develop novel statistical models for high prediction performance. When applied to help doctor to treat patients, these models allow the users to include gene or other biomarkers for predicting effectiveness of a treatment. When applied to risk management in finance, these models are capable to include an organization's or individual's ongoing finance status to predict, for example, the probability of or time to loan default.
2023–2026 Chief Investigator ($1.6M) NHMRC Ideas Grant “Computational analysis and Artificial Intelligence in brain tumour imaging: towards the augmented diagnostics of the future”. A cure for cancer cannot be achieved without improving our diagnostic accuracy with better imaging markers of disease, as well as faster and more reliable interpretation of the data. Our project for the next five years aims to meet this need by developing a reliable system for automated diagnostic augmentation in neuroradiology and neuropathology imaging by integrating computational image analysis and artificial intelligence (AI) into the diagnostic pathway and therapeutic decision-making.
2024–2026 Chief Investigator ($450,000) COCHLEAR Macquarie University joint fund “Rapid, cheap, and informative hearing screening for all”. Early identification and effective management of hearing impairment (HI) are crucial for optimal hearing health outcomes. This necessitates the development of robust screening tools that are both accurate and economical to facilitate widespread adoption, particularly in universal newborn hearing screening (UNHS) programs. Our project for the 3 years proposes to reduce the test duration to less than one minute using advanced statistical and signal-processing approaches.

Student supervision

13 PhD students supervised to completion, plus Masters and Honours students.

2026 - 2026 Supervision of a Master candidate (Tolidji Agossouvo) at UPPA
2026 - 2026 Supervision of a Master candidate (Tugdual Pennamen) at UPPA
2025 - Supervision of a Phd candidate (Soule Soilah-Dine) at UPPA
2025 - 2025 Supervision of a Master candidate (Daniel Flores) at UPPA
2023 - Supervision of a PhD student candidate (Christina Feng) at Macquarie University
2023 - Co-supervision of a PhD student candidate (Matthew Fernandez) at Macquarie University
2023 - 2024 Co-supervision of a Mres candidate (Abreen Chen) at Macquarie University
2020 - 2023 Supervision of a Phd candidate (Theo Nguyen) cotutelle between University of Pau et Pays de L'Adour and Macquarie University
2021 - 2022 Co-supervision of a Mres candidate (Isabel Li) at Macquarie University
2019 - 2022 Supervision of a Phd candidate (Bastien Mougiart) at University of Pau et Pays de L'Adour
2019 - 2021 Supervision of a Phd candidate (Alexandre Lefranc) at University of Pau et Pays de L'Adour
2018 - 2021 Supervision of a Phd candidate (Aurelien Callens) at University of Pau et Pays de L'Adour
2018 - 2021 Supervision of a Phd candidate (Sebastien Coube) at University of Pau et Pays de L'Adour
2018 - 2021 Co-Supervision of a Phd candidate (Floren Hugon) at University of Pau et Pays de L'Adour
2017 - 2021 Co-Supervision of a Phd candidate (Marc Antoine Poncet) at University of Pau et Pays de L'Adour
2017 - 2019 Co-Supervision of a Phd candidate (Sophie Defontaine) at University of Pau et Pays de L'Adour
2016 - 2019 Co-Supervision of a Phd candidate (Camilo Broc) at University of Pau et Pays de L'Adour
2016 - 2019 Co-Supervision of a Phd candidate (Matthew Sutton) at Queensland University Technology.
2013 - 2017 Co-supervision of a Phd candidate (Azam Asanjarani) at the University of Queensland.
2018 - Supervision of a Master Student in Statistics (Anglet campus from UPPA).
2017 - Supervision of a Master Student in Biostatistics (Anglet campus from UPPA).
2017 - Supervision of a Master Student regarding the MICROPOLIT project (Anglet campus from UPPA).
2016 - Co-supervision of a Master Student regarding the EOLE project (Anglet campus from UPPA).
2013 - Supervision Honours and Master students at the University of Queensland.
2010 - 2013 Supervision of a PhD candidate (Jéremie Riou) on the topic of “Multiple testing in clinical studies”.
2006 - 2009 Supervision of a PhD candidate (Moliere N'Guile Makao) on the research topic of ”Prediction of nosocomial pneumonia using multi- state models”.
2007 - 2012 Supervision of first and second year Master of Public Health trainees.
2004 - 2007 Supervision of undergraduate students in Statistics and Computer Science (IUT-STID).

Invited talks and lectures

Show the full list (49 entries)

IV49 2026 (June), Liquet B, Professeur invité, conférencier “Advanced Statistical Methods B”, organisé à University of Copenhagen (Danemark)“Workshop on Introduction to machine learning and Deep Learning”.
IV48 2026 (February), Liquet B, conférencier invité, “Overview of the Fundamentals of Deep Learning Models” 1st World Conference of Computational Neurosurgery (Sydney). (link)
IV47 2025 (Mai), Liquet B, Professeur invité, conférencier “Advanced Statistical Methods B”, organisé à University of Copenhagen (Danemark)“Workshop on Introduction to machine learning and Deep Learning”.
IV46 2025 (Mars, Liquet B, International Guest Lecture at Sebelas Maret University (Indonesia) “Data Analytics dan Machine Learning'.
IV45 2025 (February, Liquet B, International Guest Lecture at Centre of Data Science (QUT, Australia) “Workshop on Introduction to machine learning and Deep Learning”.
IV44 2024 (November), Liquet B, International Guest Lecture at the conference The 2024 International Conference on Control, Automation and Information Sciences (ICCAIS 2024), Vietnam. Organized 4-day workshop on Mathematical Engineering of Deep Learning - Foundations
IV43 2024 (April), Liquet B, International Guest Lecture at Sebelas Maret University (Indonesia) (https://indico.math.cnrs.fr/event/8848/) “Workshop on Mathematics and Deep Learning”.
IV42 2023 (April), Liquet B, Invited speaker to Journées Probabilités et Statistiqué de la fédération Margaux (https://indico.math.cnrs.fr/event/8848/) “Best Subset Selection for Principal Components Analysis and Partial Least Square models using Continuous Optimization”.
IV41 2023, Liquet B, Invited Professor, lectures for “Advanced Statistical Methods B”, organized at University of Copenhagen (Danemark)
IV40 2023 (February), Liquet B, Invited speaker to Matrix event at Melbourne University Computational Mathematics for High-Dimensional Data in Statistical Learning, talk on “Sparse group models for genomics data”.
IV40 2022, Liquet B, Invited Professor, lectures for “Advanced Statistical Methods B”, organized at University of Copenhagen (Danemark)
IV39 2021 (November), Liquet B, Invited speaker to workshop on Revolutionising water quality monitoring in the information age (https://arclpworkshop.wixsite.com/website) “A framework to infill missing data from freshwater high- frequency sensors data”.
IV38 2021 (October), Liquet B, Invited speaker to The Fourth International Conference on Statistics, Mathematics, Teaching, and Researches ICSMTR21 at Universitas Negeri Makassar (Indonesia) “Variable Selection and Dimension Reduction Methods for High Dimensional and Big-Data Set
IV37 2021 (June), Liquet B, Invited to the mathematics colloquium at School of mathematics and Physics at The university of Queensland. “Variable selection and dimension reduction methods for high dimensional and big-data set”.https://smp.uq.edu.au/event/session/10625
IV36 2021 (June), Liquet B, Invited speaker to virtual event Applications du Bayesian Unified Group of Statisticians AppliBUGS (http://genome.jouy.inra.fr/applibugs/applibugs.welcome.html) “ Sparse group models for leveraging pleiotropy effects from GWAS.”
IV35 2020 (December), Liquet B, Invited speaker to joint virtual conference CFE-CMStatistics 2020 (13th International Conference of the ERCIM WG on Computational and Methodological Statistics and 14th International Conference on Computational and Financial Econometrics) `BIG-SIR: a Sliced Inverse Regression approach for massive data.”
IV34 2020 (December), Liquet B, Invited speaker to BioInfoSummer 2020 symposium of the Australia's Mathematical Sciences Institute (AMSI), Virtual conference. “Bayesian meta-analysis models for cross cancer genomic investigation of pleiotropic effects with group structure. ”
IV33 2020 (April), Liquet B, Invited Professor, lectures for “Advanced Statistical Topics in Health Research”, organized at University of Copenhagen (Danemark) on line lectures.
IV32 2019 (November), Liquet B, Invited speaker to Statistics Across Campuses colloquium 2020 “Leveraging Pleiotropy effect from genome-wide association studies using Sparse Group Models. ” Macquarie, Australia. (October 2020)
IV31 2019 (November), Liquet B, Invited speaker at Bayes On The Beach Conference “Leverage pleiotropic effects from genome-wide association studies using both frequentist and Bayesian sparse group models”, Gold Coast, Australia
IV30 2019 (September), Liquet B, Invited lectures the Short course on Stat-XP: statistics to analyse OMICs data and characterise the exposome, London, GB
IV28 2019 (June), Liquet B, Invited speaker at International workshop on “Perspectives On High-dimensional Data Analysis” (https://indico.uu.se/event/526/) Uppsala University in Sweden
IV27 2019 (May), Liquet B, Invited Professor, seminar and lectures for “Advanced Statistical Topics in Health Research”, organized at University of Copenhagen (Danemark).
IV26 2019 (March), Invited Professor (one week) Seminar and running a workshop on “Statistical methods for high dimensional and Massive Data”, Sebelas Maret University at Solo (Indonesia)
IV25 2018 (October), Liquet B, Invited speaker at “1er événement DATA & IA à Pau les 28, 29 et 30 septembre 2018”, organized at Pau.
IV24 2018 (September), Liquet B, Invited speaker at “CIBB2018: Computational Intelligence methods for Bioinformatics and Biostatistics”, organized at Caparica (Portugal) https://eventos.fct.unl.pt/cibb2018/pages/program-invited.
IV23 2018 (July), Liquet B, Invited Professor, seminar and lectures on “Statistical methods for high dimensional data”, organized at the University of Victoria (Canada).
IV22 2018 (June), Liquet B, Invited speaker at “Workshop in honor of Daniel Commenges' 70th birthday”, organized at Bordeaux University (Bordeaux).
IV21 2018 (May), Liquet B, Invited Professor, seminar and lectures on “Statistical methods for analyzing Big-DATA”, organized at the University of Copenhagen (Danemark).
IV20 2018 (April), Liquet B, Invited speaker at International workshop on “Perspectives On High-dimensional Data Analysis” (http://hdda-viii.uca.ma)
IV19 2018 (Avril), Liquet B, Invited lectures the Short course on “Stat-XP: statistics to analyse OMICs data and characterise the exposome”, organized Imperial College London, England.
IV18 2017 (December), Liquet B. Organiser of a Contributed Session at the 2017 IASC-ARS/NZSA Conference. “Statistical Methods for the Analysis of High-Dimensional and Massive Data”, New Zealand.
IV17 2017 (October), Liquet B. Organiser of a Contributed Session at the 2017 WANRM workshop “Mathematical Methods for Modelling Natural Resources”, The University of Queensland (Australia).
IV16 2017 (October), Invited Professor (one week) Seminar and running a workshop on “Reproducible workflow analysis using R”, Sebelas Maret University at Solo (Indonesia)
IV15 2016 (December), Lafaye de Micheaux P., Liquet B. and Riou J., “Type-II generalized family-wise error rate formulas with application to sample size determination”, 9th International Conference of the ERCIM (European Research Consortium for Informatics and Mathematics) Spain.
IV14 2016 (December), “Bayesian Variable Selection Regression Of Multivariate Responses For Group Data”, Invited to the Big Bayes Session at the Australian Statistical Conference.
IV13 2016 (December), “Statistical Methods For Analysing High-Dimensional Data and Massive Data”, International Conference on Mathematics : Education, Theory, and Application (ICMETA). Universitas Sebelas Maret (Indonesia).
IV12 2016 (December), Two invited lectures: “A tutorial for penalized regression models” and “A tutorial for PLS and Bayesian Variable Selection”. Short course on “Stat-XP: statistics to analyse OMICs data and characterise the exposome”, organized Imperial College London, England.
IV11 2016 (December) Sutton M., Liquet B., and ThiÃªbaut R. “Sparse Group Subgroup PLS for Genomics”. International Symposium for Big Data Visualisation Analytics (BDVA) Sydney.
IV10 2016 (October) Lafaye de Micheaux P., Liquet B. and Sutton M., “A Unified Regularized Group PLS Algorithm Scalable to Big Data”, journée STAtistique de Rennes (jSTAR), 13rd edition on Big Data, Rennes, France.
IV9 2016 (July) Dimension Reduction approaches for BIG-DATA. Workshop on Big-Data organized by ACEMS at Queensland University of Technology.
IV8 2016 (February) Statistical Methods for Analysising High-dimensional Data and Massive Data. Lecture at the University of Melbourne.
IV7 2015 (December), Multivariate approaches: Dimension reduction for big data sets. Short course on “Statistical approaches to characterize the exposome: Overview and Perspective”, organized Imperial College London, England.
IV6 2014 (December), Multivariate approaches: Dimension reduction for big data sets. Short course on “Statistical approaches to characterize the exposome: Overview and Perspective”, organized Imperial College London, England.
IV5 2014 (November), R2GUESS: A Graphic Processing Unit-Based R Package for Bayesian Variable Selection Regression of Multivariate Responses, Bayes on the Beach conference organized in Gold Coast, Australia.
IV4 2013 (October), Multi-state models and nosocomial infections. Workshop on “dynamic predictions for repeated markers and repeated events: models and validation in cancer”, organized in Bordeaux, France.
IV3 2013 (April), Strategies to analyse 'omics' data: From standard T-tests in genomics to a more general Bayesian approach. Workshop Statlearn'13. Bordeaux University (France).
IV2 2009 (November), Choice of estimators based on different observations: Modified AIC and LCV criteria. Colloque CRM-ISM-GERAD de Statistique. McGill University (Canada).
IV1 2009 (October), Choix d'estimateurs basés sur des observations différentes. GDR “Statistique et Santé”, Université Paris-Descartes.