DATA AND AI ARE CHANGING THE WAY ORGANIZATIONS THINK, DECIDE, AND ORGANIZE. IT’S TIME HUMANITIES, MANAGEMENT AND SOCIAL SCIENCES GET INVOLVED.
Metalab

NEWS

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.

EVENTS

[ARTICLE] This paper establishes risk bounds for Regularized Empirical Risk Minimizers (RERM) under both subgaussian and heavy-tailed assumptions, introducing an alternative minmax MOM procedure for robust estimation in the presence of outliers.

by Guillaume Lecué (ESSEC Business School), Geoffrey Chinot, Matthieu Lerasle

We establish risk bounds for Regularized Empirical Risk Minimizers (RERM) when the loss is Lipschitz and convex and the regularization function is a norm. In a first part, we obtain these results in the i.i.d. setup under subgaussian assumptions on the design. In a second part, a more general framework where the design might have heavier tails and data may be corrupted by outliers both in the design and the response variables is considered. In this situation, RERM performs poorly in general. We analyse an alternative procedure based on median-of-means principles and called “minmax MOM”. We show optimal subgaussian deviation rates for these estimators in the relaxed setting. The main results are meta-theorems allowing a wide-range of applications to various problems in learning theory. To show a non-exhaustive sample of these potential applications, it is applied to classification problems with logistic loss functions regularized by LASSO and SLOPE, to regression problems with Huber loss regularized by Group LASSO and Total Variation. Another advantage of the minmax MOM formulation is that it suggests a systematic way to slightly modify descent based algorithms used in high-dimensional statistics to make them robust to outliers. We illustrate this principle in a Simulations section where a “ minmax MOM” version of classical proximal descent algorithms are turned into robust to outliers algorithms.

[Please read the research paper here]

[ARTICLE] This paper leverages advanced techniques from empirical process theory and Statistical Learning Theory to provide precise statistical analysis and guarantees for Semi-Definite Programming (SDP) estimators across various machine learning problems.

by Guillaume Lecué (ESSEC Business School), Stéphane Chrétien, Mihai Cucuringu, Lucie Neirac

Many statistical learning problems have recently been shown to be amenable to Semi-Definite Programming (SDP), with community detection and clustering in Gaussian mixture models as the most striking instances Javanmard et al. (2016). Given the growing range of applications of SDP-based techniques to machine learning problems, and the rapid progress in the design of efficient algorithms for solving SDPs, an intriguing question is to understand how the recent advances from empirical process theory and Statistical Learning Theory can be leveraged for providing a precise statistical analysis of SDP estimators. In the present paper, we borrow cutting edge techniques and concepts from the Learning Theory literature, such as fixed point equations and excess risk curvature arguments, which yield general estimation and prediction results for a wide class of SDP estimators. From this perspective, we revisit some classical results in community detection from Guédon and Vershynin (2016) and Fei and Chen (2019), and we obtain statistical guarantees for SDP estimators used in signed clustering, angular group synchronization (for both multiplicative and additive models) and MAX-CUT. Our theoretical findings are complemented by numerical experiments for each of the three problems considered, showcasing the competitiveness of the SDP estimators.

[Please read the research paper here]

[ARTICLE] This paper constructs a robust alternative to cross-validation for hyperparameter tuning and model selection using a median-of-means principle.

by Guillaume Lecué (ESSEC Business School), Joon Kwon, Matthieu Lerasle

Hyperparameter tuning and model selection are important steps in machine learning. Unfortunately, classical hyperparameter calibration and model selection procedures are sensitive to outliers and heavy-tailed data. In this work, we construct a selection procedure which can be seen as a robust alternative to cross-validation and is based on a median-of-means principle. Using this procedure, we also build an ensemble method which, trained with algorithms and corrupted heavy-tailed data, selects an algorithm, trains it with a large uncorrupted subsample and automatically tunes its hyperparameters. In particular, the approach can transform any procedure into a robust to outliers and to heavy-tailed data procedure while tuning automatically its hyperparameters.

[Please read the research paper here]

[ARTICLE] This paper proposes a new parameter constraint for non-Gaussian mixture models using a small dictionary of elements.

by Pierre Alquier (ESSEC Business School), Léna Carel

Mixture models are among the most popular tools for clustering. However, when the dimension and the number of clusters is large, the estimation of the clusters become challenging, as well as their interpretation. Restriction on the parameters can be used to reduce the dimension. An example is given by mixture of factor analyzers for Gaussian mixtures. The extension of MFA to non-Gaussian mixtures is not straightforward. We propose a new constraint for parameters in non-Gaussian mixture model: the K components parameters are combinations of elements from a small dictionary, say H elements, with H≪K. Including a nonnegative matrix factorization (NMF) in the EM algorithm allows us to simultaneously estimate the dictionary and the parameters of the mixture. We propose the acronym NMF-EM for this algorithm, implemented in the R package nmfem. This original approach is motivated by passengers clustering from ticketing data: we apply NMF-EM to data from two Transdev public transport networks. In this case, the words are easily interpreted as typical slots in a timetable.

[Please read the research paper here]

[ARTICLE] This paper proposes a meta-strategy for online learning methods like the online gradient algorithm (OGA) and exponentially weighted aggregation (EWA) to learn tuning parameters from past tasks.

by Pierre Alquier (ESSEC Business School), Dimitri Meunier

Online learning methods, similar to the online gradient algorithm (OGA) and exponentially weighted aggregation (EWA), often depend on tuning parameters that are difficult to set in practice. We consider an online meta-learning scenario, and we propose a meta-strategy to learn these parameters from past tasks. Our strategy is based on the minimization of a regret bound. It allows us to learn the initialization and the step size in OGA with guarantees. It also allows us to learn the prior or the learning rate in EWA. We provide a regret analysis of the strategy. It allows to identify settings where meta-learning indeed improves on learning each task in isolation.

[Please read the research paper here]

[ARTICLE] This paper proposes and evaluates a comprehensible bivariate, perceptually optimized visualization scheme for high-dimensional data.

by Adalbert Wilhelm (ESSEC Business School), André Müller, Ludwig Lausser, Timo Ropinski, Matthias Platzer, Heiko Neumann, Hans A. Kestler

Visualising data as diagrams using visual attributes such as colour, shape, size, and orientation is challenging. In particular, large data sets demand graphical display as an essential step in the analysis. In order to achieve comprehension often different attributes need to be displayed simultaneously. In this work a comprehensible bivariate, perceptually optimised visualisation scheme for high-dimensional data is proposed and evaluated. It can be used to show fold changes together with confidence values within a single diagram. The visualisation scheme consists of two parts: a uniform, symmetric, two-sided colour scale and a patch grid representation. Evaluation of uniformity and symmetry of the two-sided colour scale was performed in comparison to a standard RGB scale by twenty-five observers. Furthermore, the readability of the generated map was validated and compared to a bivariate heat map scheme.

[Please read the research paper here]

[ARTICLE] This paper presents a novel Optimizer algorithm for automatic apartment layout generation.

by Florian Brun (ESSEC Business School), Graziella Laignel, Nicolas Pozin, Xavier Geffrier, Loukas Delevaux, Bastien Dolla

We present a novel approach for automatic apartment layout generation. Given a polygonal apartment envelope and a list of rooms with associated area, our so-called Optimizer algorithm generates several floor plans aiming at both architectural and functional constraints.

To do so, Optimizer discretizes the floor space into a grid according to architectural constraints and reduces the problem to a cell assignment which is solved through a coupled constraint programming - genetic optimization approach.

Obtained results demonstrate the feasibility of our approach, customized plans are architecturally and functionally valid, they are mostly generated in about 1 min.

[Please read the research paper here]

[ARTICLE] This research demonstrates that machine learning algorithms, particularly random forest classifiers, can effectively classify borrowers into credit categories using readily available data.

by Apostolos Ampountolas (ESSEC Business School), Titus Nyarko Nde, Paresh Date, Corina Constantinescu

In micro-lending markets, lack of recorded credit history is a significant impediment to assessing individual borrowers’ creditworthiness and therefore deciding fair interest rates. This research compares various machine learning algorithms on real micro-lending data to test their efficacy at classifying borrowers into various credit categories. We demonstrate that off-the-shelf multi-class classifiers such as random forest algorithms can perform this task very well, using readily available data about customers (such as age, occupation, and location). This presents inexpensive and reliable means to micro-lending institutions around the developing world with which to assess creditworthiness in the absence of credit history or central credit databases.

[Please read the research paper here]

[ARTICLE] This study presents a machine learning framework incorporating text analysis of keyword series to improve hotel demand predictions.

by Apostolos Ampountolas (ESSEC Business School), Mark P. Legg

Purpose
This study aims to predict hotel demand through text analysis by investigating keyword series to increase demand predictions’ precision. To do so, this paper presents a framework for modeling hotel demand that incorporates machine learning techniques.

Design/methodology/approach
The empirical forecasting is conducted by introducing a segmented machine learning approach of leveraging hierarchical clustering tied to machine learning and deep learning techniques. These features allow the model to yield more precise estimates. This study evaluates an extensive range of social media–derived words with the most significant probability of gradually establishing an understanding of an optimal outcome. Analyzes were performed on a major hotel chain in an urban market setting within the USA.

Findings
The findings indicate that while traditional methods, being the naïve approach and ARIMA models, struggled with forecasting accuracy, segmented boosting methods (XGBoost) leveraging social media predict hotel occupancy with greater precision for all examined time horizons. Additionally, the segmented learning approach improved the forecasts’ stability and robustness while mitigating common overfitting issues within a highly dimensional data set.

Research limitations/implications
Incorporating social media into a segmented learning framework can augment the current generation of forecasting methods’ accuracy. Moreover, the segmented learning approach mitigates the negative effects of market shifts (e.g. COVID-19) that can reduce in-production forecasts’ life-cycles. The ability to be more robust to market deviations will allow hospitality firms to minimize development time.

Originality/value
The results are expected to generate insights by providing revenue managers with an instrument for predicting demand.

[Please read the research paper here]

[ARTICLE] This paper proposes a generic data model for prosopographic databases, enhancing their quality by integrating concepts like uncertainty, reliability, accuracy, and completeness.

by Isabelle Wattiau, Jacky Akoka (ESSEC Business School), Stéphane Lamasse, Cédric du Mouza

Prosopographic databases, which allow the study of social groups through their bibliography, are used today by a significant number of historians. Computerization has allowed intensive and large-scale exploitation of these databases. The modeling of these proposopographic databases has given rise to several data models. An important problem is to ensure a level of quality of the stored information. In this article , we propose a generic data model allowing to describe most of the existing prosopographic databases and to enrich them by integrating several quality concepts such as uncertainty, reliability, accuracy or completeness.

[Please read the research paper here]

Founded in 2020 by ESSEC Business School, The Institute for AI and Business for Artificial Intelligence, Data and Society helps organizations navigate and better understand the social, economic, cultural, and ethical impacts of AI and data

metalab@essec.edu

Learn more about the Institute for AI and Business

copyright © 2026 Institute for AI and Business

↓
Résumé de la politique de confidentialité

Ce site utilise des cookies afin que nous puissions vous fournir la meilleure expérience utilisateur possible. Les informations sur les cookies sont stockées dans votre navigateur et remplissent des fonctions telles que vous reconnaître lorsque vous revenez sur notre site Web et aider notre équipe à comprendre les sections du site que vous trouvez les plus intéressantes et utiles.