Online Machine Learning in Montpellier, Theory & Practice Our objective is to construct well-calibrated prediction sets for a time-to-event outcome subject to right-censoring with guaranteed coverage. Our approach is inspired by modern conformal inference literature, in that, unlike classical frameworks, we obviate the need for a well-specified parametric or semi-parametric survival model to accomplish our goal. In contrast to existing conformal prediction methods for survival data, which restrict censoring to be of Type I, whereby potential censoring times are assumed to be fully observed on all units in both training and validation samples, we consider the more common right-censoring setting in which either only the censoring time or only the event time of primary interest is directly observed, whichever comes first. Under a standard conditional independence assumption between the potential survival and censoring times given covariates, we propose and analyze two methods to construct valid and efficient lower predictive bounds for the survival time of a future observation. The proposed methods build upon modern semiparametric efficiency theory for censored data, in that the first approach incorporates inverse-probability-of-censoring weighting (IPCW), while the second approach is based on augmented-inverse-probability-of-censoring weighting (AIPCW). For both methods, we formally establish asymptotic coverage guarantees, and demonstrate both via theory and empirical experiments that AIPCW substantially improves efficiency over IPCW in the sense that its coverage error bound is of second-order mixed bias type, that is doubly robust, and therefore guaranteed to be asymptotically negligible relative to the coverage error of IPCW. Machine Learning in Montpellier, Theory & Practice

Campus St Priest Rebecca Farina Our objective is to construct well-calibrated prediction sets for a time-to-event outcome subject to right-censoring with guaranteed coverage. Our approach is inspired by modern conformal inference literature, in that, unlike classical frameworks, we obviate the need for a well-specified parametric or semi-parametric survival model to accomplish our goal. In contrast to existing conformal prediction methods for survival data, which restrict censoring to be of Type I, whereby potential censoring times are assumed to be fully observed on all units in both training and validation samples, we consider the more common right-censoring setting in which either only the censoring time or only the event time of primary interest is directly observed, whichever comes first. Under a standard conditional independence assumption between the potential survival and censoring times given covariates, we propose and analyze two methods to construct valid and efficient lower predictive bounds for the survival time of a future observation. The proposed methods build upon modern semiparametric efficiency theory for censored data, in that the first approach incorporates inverse-probability-of-censoring weighting (IPCW), while the second approach is based on augmented-inverse-probability-of-censoring weighting (AIPCW). For both methods, we formally establish asymptotic coverage guarantees, and demonstrate both via theory and empirical experiments that AIPCW substantially improves efficiency over IPCW in the sense that its coverage error bound is of second-order mixed bias type, that is doubly robust, and therefore guaranteed to be asymptotically negligible relative to the coverage error of IPCW. LabéliséHallesIA

Campus St Priest Machine Learning in Montpellier, Theory & Practice Jim Smit Despite several randomised controlled trials (RCTs) on the use of adjuvant treatment with corticosteroids in patients with community-acquired pneumonia (CAP), the effect of this intervention on mortality remains controversial. We aimed to evaluate heterogeneity of treatment effect (HTE) of adjuvant treatment with corticosteroids on 30-day mortality in patients with CAP. [...] LabéliséHallesIA

Laboratoire d'Informatique, de Robotique et de Microélectronique de Montpellier (LIRMM) Centre de l'Imaginaire Scientifique et Technique Le Centre de l'Imaginaire Scientifique et Technique (CIST), le Défi Clé de la Région Occitanie-Pyrénées-Méditerranée «Robotique centrée sur l’humain », le Laboratoire d’Informatique, de Robotique et de Microélectronique de Montpellier (LIRMM), l’Université de Montpellier, le CNRS, ont décidé de proposer un temps convivial , passionnant et hors normes, qui permette à tous de se rencontrer et d’échanger mais aussi de proposer et soutenir des actions efficaces dans le domaine de la relation «science et société». Vous êtes impliqué ou concerné professionnellement par les domaines de la robotique, l’intelligence artificielle, le traitement de données et autres domaines connexes (informatique, mathématiques, mécanique...) ? Qu’il s’agisse de la recherche, de l’enseignement, de l’innovation et la production, de la communication, de la réflexion éthique et durable… cette initiative régionale localisée à Montpellier vous est destinée. Avec la révolution technologique, économique et sociétale en cours, une multitude d’initiatives naissent qui couvrent un vaste spectre d’approches répondant à de nombreux besoins. Bien sûr, certaines émanent de laboratoires et organismes de recherche, universités et autres structures d'enseignement, d’autres des milieux industriels, des médias et des associations de médiation scientifique et technique qui traitent tous les questions vives de la société et valorisent les métiers scientifiques et l’innovation. Inscription gratuite mais obligatoire LabéliséHallesIA

Room 02.022, Building 5, St Priest campus Machine Learning in Montpellier, Theory & Practice In the first part of the talk, I will present some reflections on the purpose of prediction sets and the role that statistics can play in forming useful prediction sets. In particular, I will discuss how prediction sets fit into a decision making pipeline and the different kinds of decisions one may make using a prediction set. In the second part of the talk, I will describe a particular statistically motivated set-generating procedure for the classification setting called clustered conformal prediction, which gives all classes an equal chance of being correctly included in the prediction set (“class-conditional coverage”). This procedure can be useful in situations where it is important to identify instances of all classes, even the rare ones. We demonstrate the performance of this method on ImageNet and other image classification datasets. Machine Learning in Montpellier, Theory & Practice

Campus St Priest (860 Rue Saint Priest 34095 Montpellier Cedex 5), bat. 5, Room: 02.124 Machine Learning in Montpellier, Theory & Practice Tiffany Ding (UC Berckley) In the first part of the talk, I will present some reflections on the purpose of prediction sets and the role that statistics can play in forming useful prediction sets. In particular, I will discuss how prediction sets fit into a decision making pipeline and the different kinds of decisions one may make using a prediction set. In the second part of the talk, I will describe a particular statistically motivated set-generating procedure for the classification setting called clustered conformal prediction, which gives all classes an equal chance of being correctly included in the prediction set (“class-conditional coverage”). This procedure can be useful in situations where it is important to identify instances of all classes, even the rare ones. We demonstrate the performance of this method on ImageNet and other image classification datasets. LabéliséHallesIA

Faculté de Droit et de Science politique - Campus Droit, centre ville - Amphi C Paul Valéry - Bat. 1 - 39 rue de l'Université, Montpellier Laboratoire Innovation Communication et Marché (LICeM) en partenariat avec l’Association Française des Juristes d’Entreprise (AFJE) Cette manifestation annuelle a pour objectif d’assurer une veille législative et jurisprudentielle dans les différents et nombreux domaines du Droit de l’internet. La journée se décline en deux temps : • matin (sous la présidence de A. Robin, LICeM) : Veille dans le domaine de la responsabilité des prestataires techniques, droit de la propriété intellectuelle, droit de la consommation, droit du travail, protection des données personnelles. • après-midi (sous la présidence de A.-E. Rousseau, AFJE) : Table-ronde autour du thème de « l’utilisation des SIA générative dans la pratique des juristes d’entreprise ».

Online Machine Learning in Montpellier, Theory & Practice We study the problem of distributional matrix completion: Given a sparsely observed matrix of empirical distributions, we seek to impute the true distributions associated with both observed and unobserved matrix entries. This is a generalization of traditional matrix completion where the observations per matrix entry are scalar-valued. To do so, we utilize tools from optimal transport to generalize the nearest neighbors method to the distributional setting. Under a suitable latent factor model on probability distributions, we establish that our method recovers the distributions in the Wasserstein metric. We demonstrate through simulations that our method (i) provides better distributional estimates for an entry compared to using observed samples for that entry alone, (ii) yields accurate estimates of distributional quantities such as standard deviation and value-at-risk, and (iii) inherently supports heteroscedastic distributions. In addition, we demonstrate our method on a real-world quarterly earnings predictions dataset. We also prove novel asymptotic results for Wasserstein barycenters over one-dimensional distributions. Machine Learning in Montpellier, Theory & Practice

Campus St Priest (860 Rue Saint Priest 34095 Montpellier Cedex 5), bat. 5, Room: 02.124 Machine Learning in Montpellier, Theory & Practice Jacob Feitelberg We study the problem of distributional matrix completion: Given a sparsely observed matrix of empirical distributions, we seek to impute the true distributions associated with both observed and unobserved matrix entries. This is a generalization of traditional matrix completion where the observations per matrix entry are scalar-valued. To do so, we utilize tools from optimal transport to generalize the nearest neighbors method to the distributional setting. Under a suitable latent factor model on probability distributions, we establish that our method recovers the distributions in the Wasserstein metric. We demonstrate through simulations that our method (i) provides better distributional estimates for an entry compared to using observed samples for that entry alone, (ii) yields accurate estimates of distributional quantities such as standard deviation and value-at-risk, and (iii) inherently supports heteroscedastic distributions. In addition, we demonstrate our method on a real-world quarterly earnings predictions dataset. We also prove novel asymptotic results for Wasserstein barycenters over one-dimensional distributions. LabéliséHallesIA

Inria Montpellier, St-Priest Campus, Building 5, Room 02/022 Machine Learning in Montpellier, Theory & Practice Toby Hocking Data.table is one of the most efficient open-source in-memory data manipulation packages available today. First released to CRAN by Matt Dowle in 2006, it continues to grow in popularity, and now over 1500 other CRAN packages depend on data.table. This talk will start with data reading from CSV, discuss basic and advanced data manipulation topics, and finally will end with a discussion about how you can contribute to data.table. https://github.com/tdhock/2023-10-LatinR-data.table?tab=readme-ov-file#source-files-for-latinr-datatable-tutorial-slides LabéliséHallesIA

Online Machine Learning in Montpellier, Theory & Practice Session organised by IDESP. Machine Learning in Montpellier, Theory & Practice

Room 02.022, Building 5, St Priest campus Machine Learning in Montpellier, Theory & Practice Data.table is one of the most efficient open-source in-memory data manipulation packages available today. First released to CRAN by Matt Dowle in 2006, it continues to grow in popularity, and now over 1500 other CRAN packages depend on data.table. This talk will start with data reading from CSV, discuss basic and advanced data manipulation topics, and finally will end with a discussion about how you can contribute to data.table. https://isdm.umontpellier.fr/github.com/tdhock/2023-10-LatinR-data.table?tab=readme-ov-file#source-files-for-latinr-datatable-tutorial-slides Machine Learning in Montpellier, Theory & Practice

Salle Ile aux contes – Mezzanine – Médiathèque Emile Zola (218 Bd. de l’Aéroport international, 34000 Montpellier) Madalina Croitoru et Ganesh Gowrishankar Venez découvrir QT, Le robot conteur, pour un moment de lecture unique! Cet atelier sera animé par Madalina Croitoru et Ganesh Gowrishankar et destiné aux enfants de 5 à 7 ans. Sur inscription à la banque d’accueil jeunesse au 1er étage ou au 04 99 06 27 34

Salle des conseils P Raynaud, au 2ème étage, du bâtiment 11 dit le château, 2 Place Pierre Viala Campus La Gaillarde SEmantic web SeminAr MontpEllier Pierre BISQUERT (INRAE IATE) Abstract: One of the major benefits of symbolic AI is explainability. When new knowledge is obtained via a reasoning process, it is possible to determine precisely the elements of the knowledge base that yield this knowledge. Typically, one would use a SAT solver to compute the explanations. However, SAT-solving is computationally expensive, and as the knowledge base grows, the time required increases exponentially. In this talk, we will 1) discuss the notion(s) of explanation of a query in the context of a (Datalog) knowledge base, then 2) we will present a method to optimise the time used by the SAT solver (by creating a hypergraph representing the grounded knowledge base and pruning the nodes that are not reachable from the fact that we want to explain), and finally 3) we will see its implementation in the context of InteGraal, a tool for reasoning over heterogeneous and federated data sources. LabéliséHallesIA

Salle des conseils P Raynaud, au 2ème étage, du bâtiment 11 dit le château, 2 Place Pierre Viala Campus La Gaillarde SEmantic web SeminAr MontpEllier Pierre BISQUERT (INRAE IATE) One of the major benefits of symbolic AI is explainability. When new knowledge is obtained via a reasoning process, it is possible to determine precisely the elements of the knowledge base that yield this knowledge. Typically, one would use a SAT solver to compute the explanations. However, SAT-solving is computationally expensive, and as the knowledge base grows, the time required increases exponentially. LabéliséHallesIA

Online Machine Learning in Montpellier, Theory & Practice We introduce efficient plug-in (EP) learning, a novel framework for the estimation of heterogeneous causal contrasts, such as the conditional average treatment effect and conditional relative risk. The EP-learning framework enjoys the same oracle-efficiency as Neyman-orthogonal learning strategies, such as DR-learning and R-learning, while addressing some of their primary drawbacks, including that (i) their practical applicability can be hindered by loss function non-convexity; and (ii) they may suffer from poor performance and instability due to inverse probability weighting and pseudo-outcomes that violate bounds. To avoid these drawbacks, EP-learner constructs an efficient plug-in estimator of the population risk function for the causal contrast, thereby inheriting the stability and robustness properties of plug-in estimation strategies like T-learning. Under reasonable conditions, EP-learners based on empirical risk minimization are oracle-efficient, exhibiting asymptotic equivalence to the minimizer of an oracle-efficient one-step debiased estimator of the population risk function. In simulation experiments, we illustrate that EP-learners of the conditional average treatment effect and conditional relative risk outperform state-of-the-art competitors, including T-learner, R-learner, and DR-learner. Open-source implementations of the proposed methods are available in our R package hte3. Machine Learning in Montpellier, Theory & Practice

Room Nadir, Maison de la Télédétection (500 rue Jean François Breton) Machine Learning in Montpellier, Theory & Practice - Lars Van der Laan We introduce efficient plug-in (EP) learning, a novel framework for the estimation of heterogeneous causal contrasts, such as the conditional average treatment effect and conditional relative risk. The EP-learning framework enjoys the same oracle-efficiency as Neyman-orthogonal learning strategies, such as DR-learning and R-learning, while addressing some of their primary drawbacks, including that (i) their practical applicability can be hindered by loss function non-convexity; and (ii) they may suffer from poor performance and instability due to inverse probability weighting and pseudo-outcomes that violate bounds. To avoid these drawbacks, EP-learner constructs an efficient plug-in estimator of the population risk function for the causal contrast, thereby inheriting the stability and robustness properties of plug-in estimation strategies like T-learning. Under reasonable conditions, EP-learners based on empirical risk minimization are oracle-efficient, exhibiting asymptotic equivalence to the minimizer of an oracle-efficient one-step debiased estimator of the population risk function. In simulation experiments, we illustrate that EP-learners of the conditional average treatment effect and conditional relative risk outperform state-of-the-art competitors, including T-learner, R-learner, and DR-learner. Open-source implementations of the proposed methods are available in our R package hte3. LabéliséHallesIA

Room Nadir, Maison de la Télédétection, Agropolis campus Machine Learning in Montpellier, Theory & Practice - Martin Atzmüller (DFKI / Osnabrück University) In many applications, modeling complex data is of utmost importance, requiring the use of advanced machine learning models and approaches. However, in many domains users require insight into models and/or their decisions, which is not necessarily provided by the respective models per se. Explainable and interpretable learning approaches can facilitate such insights for making sense of models and decisions. The talk presents examples of such approaches in complex modeling domains, including interpretable as well as explainable deep-learning-based methods, and a neuro-symbolic architecture including domain knowledge for facilitating explainability. Machine Learning in Montpellier, Theory & Practice

Room Nadir, Maison de la Télédétection (500 rue Jean François Breton) Machine Learning in Montpellier, Theory & Practice - Martin Atzmüller, Scientific Director at DFKI and Full Professor at Osnabrück University (Germany) In many applications, modeling complex data is of utmost importance, requiring the use of advanced machine learning models and approaches. However, in many domains users require insight into models and/or their decisions, which is not necessarily provided by the respective models per se. Explainable and interpretable learning approaches can facilitate such insights for making sense of models and decisions. The talk presents examples of such approaches in complex modeling domains, including interpretable as well as explainable deep-learning-based methods, and a neuro-symbolic architecture including domain knowledge for facilitating explainability. LabéliséHallesIA

Campus Triolet Polytech, Salle du Conseil (1er étage) Montpellier BioInfo - Céline Mandier (ISDM) La bioinformatique repose sur des analyses complexes et des volumes de données conséquents. Adopter de bonnes pratiques permet d’assurer fiabilité, reproductibilité et efficacité. Accueil petit dej à partir de 9h00 LabéliséHallesIA

Inria Montpellier, St-Priest Campus, Building 5, Room 02/124 Machine Learning in Montpellier, Theory & Practice - Vincent Mallet Learning on the 3D structure of biomolecules resulted in several breakthroughs, for instance the de-novo design of potent binders or functional luciferases, eventually leading to a Nobel Prize. Most of this work relies on modeling the structure of protein as graphs, with well-established tools. IA&Experts, LabéliséHallesIA

Room 02.124, Building 5, St Priest campus Machine Learning in Montpellier, Theory & Practice - Vincent Mallet (Ecole des Mines de Paris) Learning on the 3D structure of biomolecules resulted in several breakthroughs, for instance the de-novo design of potent binders or functional luciferases, eventually leading to a Nobel Prize. Most of this work relies on modeling the structure of protein as graphs, with well-established tools. In my talk I will start by introducing the topic and reviewing recent methods. I will then present our recent results on the coordinated use of graphs, sequence and surface representations to model protein structure with machine learning. We show that despite disappointing results in isolation, surface methods used in addition to others display a synergistic effect. I will also discuss ongoing work to enhance this method, as well as apply it to homologs mining and binder design. About the presenter: Vincent Mallet is a researcher working on modeling the structure of biomolecules with machine learning, with applications in structural biology and drug design. His two main lines of work are surface-based representations for proteins and graph-based representations for RNA. He is part of the CBIO team, in Mines Paris PSL and Institut Curie. Previously, he was a postdoc with Maks Ovsjanikov working on geometric deep learning. He did his PhD with Jean-Philippe Vert and Michael Nilges on equivariant methods and drug design. Machine Learning in Montpellier, Theory & Practice

Room 02.124, Building 5, St Priest campus Machine Learning in Montpellier, Theory & Practice - François Petit (Inserm) We developed a mathematical setup inspired by Buyse's generalized pairwise comparisons to define the notion of an optimal individualized treatment rule (ITR) in the presence of a prioritized outcomes in a randomized controlled trial, terming such an ITR pairwise optimal. We present two approaches to estimate pairwise optimal ITRs. The first is a variant of the k-nearest neighbors algorithm. The second is a meta-learner based on a randomized bagging scheme, allowing the use of any classification algorithm for constructing an ITR. We study the behavior of these estimation schemes from a theoretical standpoint and through Monte Carlo simulations and illustrate their use on trials data. Machine Learning in Montpellier, Theory & Practice

Online Machine Learning in Montpellier, Theory & Practice - Shubhankar Mohapatra (University of Waterloo) Many critical AI applications, such as personalized health assistants and social network recommender systems, rely on learning from private data. Differential privacy has become the gold standard for training models on sensitive data, gaining widespread adoption in industry and government. This growing adoption has fueled research in differentially private learning, yet significant challenges remain in its deployment. In this talk, I will discuss several challenges that arise when integrating differential privacy into an end-to-end learning pipeline, from data collection to model training. I will particularly focus on the difficulties caused by inconsistencies in private datasets, such as typos and missing values. Correcting these errors is especially challenging when direct access to the raw data is restricted due to privacy constraints. I will present two recent works addressing this issue. First, I will show how leveraging correlations and dependencies in the data lets us privately quantify inconsistencies, helping estimate data quality and the effort needed for data repair. Second, I will examine the impact of missing values in private learning and introduce simple yet effective techniques for generating differentially private synthetic data to mitigate these effects. Machine Learning in Montpellier, Theory & Practice

Amphithéâtre A (Bat. A) au campus de l’UFR STAPS de Montpellier Zaineb AJRA (EuroMov Digital Health in Motion) - Thèse AXIAUM (financement ANR) Sous la direction de Jacky MONTMAIN et Stéphane PERREY et l’encadrement de Binbin XU IA&Recherche, LabéliséHallesIA

Room 02.124, Building 5, St Priest campus Machine Learning in Montpellier, Theory & Practice - Batiste Le Bars (Inria) Conformal Prediction has recently been considered as one of the state-of-art technique to construct distribution-free prediction sets satisfying probabilistic coverage guarantees. In this talk, we study the question of volume optimality in split conformal regression. Using the fact that the calibration step can be seen as an empirical volume minimization problem, we first derive a finite-sample upper-bound on the excess volume loss of the interval returned by the classical split method. This quantity measures the difference in length between the interval obtained with the split method and the shortest oracle prediction interval. Then, we introduce EffOrt, a methodology that modifies the learning step so that the base prediction function minimizes the length of the returned intervals. In particular, our theoretical analysis of the excess volume loss of the prediction sets produced by EffOrt reveals the links between the learning and calibration steps, and notably the impact of the function class of the base predictor. We also introduce Ad-EffOrt, an extension of the previous method, which produces intervals whose size adapts to the value of the covariate. Machine Learning in Montpellier, Theory & Practice

Textmine Le groupe de travail Textmine vous invite à une journée exceptionnelle de rencontres et d’échanges entre acteurs du public et du privé autour des IA appliquées aux textes. Cet événement se tiendra le 02 juin 2025 à Montpellier, dans la salle du Conseil de l’Hôtel de Métropole (50 place Zeus – 34000 Montpellier), à seulement 10 minutes à pied de la gare. Vous êtes chercheur, entreprise, startup ou acteur public ? Venez présenter vos travaux (courtes interventions ou démos) ou simplement assister aux échanges ! Dans les deux cas, l’inscription est requise.

Campus St Priest (860 Rue Saint Priest 34095 Montpellier Cedex 5), bat. 5, Room: 02.124 Machine Learning in Montpellier, Theory & Practice - Olivier Cappé (CNRS / ENS / Université PSL) Performative learning addresses the increasingly pervasive situations in which algorithmic decisions may induce changes in the data distribution as a consequence of their public deployment. We propose a novel view in which these performative effects are modelled as push-forward measures. This general framework encompasses existing models and enables novel performative gradient estimation methods, leading to more efficient and scalable learning strategies. For distribution shifts, unlike previous models which require full specification of the data distribution, we only assume knowledge of the shift operator that represents the performative changes. Focusing on classification with a linear-in-parameters performative effect, we prove the convexity of the performative risk under a new set of assumptions. We also establish a connection with adversarially robust classification by reformulating the minimization of the performative risk as a min-max variational problem. IA&Experts

Room 02.124, Building 5, St Priest campus Machine Learning in Montpellier, Theory & Practice - Olivier Cappé () Performative learning addresses the increasingly pervasive situations in which algorithmic decisions may induce changes in the data distribution as a consequence of their public deployment. We propose a novel view in which these performative effects are modelled as push-forward measures. This general framework encompasses existing models and enables novel performative gradient estimation methods, leading to more efficient and scalable learning strategies. For distribution shifts, unlike previous models which require full specification of the data distribution, we only assume knowledge of the shift operator that represents the performative changes. Focusing on classification with a linear-in-parameters performative effect, we prove the convexity of the performative risk under a new set of assumptions. We also establish a connection with adversarially robust classification by reformulating the minimization of the performative risk as a min-max variational problem. Machine Learning in Montpellier, Theory & Practice