arXiv subjects

Olivier Teste

Publications and source records attributed to Olivier Teste.

At least 19 recordsLinked to original sources

arXiv2026

Encryption-Compatible Clustered Federated Learning via Distributed Expectation-Maximization over Metadata

Clustered Federated Learning (CFL) addresses data heterogeneity in federated settings by grouping clients with similar data distributions to enable effective training. Existing methods face a trade-off between privacy preservation, communication cost, and computational efficiency. We formalize this as the CFL trilemma, according to which improving two of these dimensions comes at the expense of the third. A prominent paradigm relies on metadata (i.e., low-dimensional representations of client datasets shared with the server) to enable communication- and computation-efficient clustering. However, such approaches are not compatible with standard FL privacy-preserving mechanisms. To address this limitation, we propose FLAMECHE, which reformulates metadata-based CFL as a distributed Expectation-Maximization (EM) procedure, restricting server updates to additive operations while preserving efficiency. This design enables compatibility with practical secure FL schemes. We conducted extensive experiments on multiple datasets under various heterogeneous scenarios. Results show that FLAMECHE improves the effectiveness of client models. It enables encryption-compatible metadata-based clustering, enhancing its positioning within the CFL trilemma.

cs.LG↗

arXiv2026

Fun-TSG: A Function-Driven Multivariate Time Series Generator with Variable-Level Anomaly Labeling

Reliable evaluation of anomaly detection methods in multivariate time series remains an open challenge, largely due to the limitations of existing benchmark datasets. Current resources often lack fine-grained anomaly annotations, do not provide explicit intervariable and temporal dependencies, and offer little insight into the underlying generative mechanisms. These shortcomings hinder the development and rigorous comparison of detection models, especially those targeting interpretable and variable-specific outputs. To address this gap, we introduce Fun-TSG, a fully customizable time series generator designed to support high-quality evaluation of anomaly detection systems. Our tool enables both fully automated generation, based on randomly sampled dependency structures and anomaly types, and manual generation through user-defined equations and anomaly configurations. In both cases, it provides full transparency over the data generation process, including access to ground-truth anomaly labels at the variable and timestamp levels. Fun-TSG supports the creation of diverse, interpretable, and reproducible benchmarking scenarios, enabling fine-grained performance analysis for both classical and modern anomaly detection models.

cs.AI↗

arXiv2026

IPatch: A Multi-Resolution Transformer Architecture for Robust Time-Series Forecasting

Accurate forecasting of multivariate time series remains challenging due to the need to capture both short-term fluctuations and long-range temporal dependencies. Transformer-based models have emerged as a powerful approach, but their performance depends critically on the representation of temporal data. Traditional point-wise representations preserve individual time-step information, enabling fine-grained modeling, yet they tend to be computationally expensive and less effective at modeling broader contextual dependencies, limiting their scalability to long sequences. Patch-wise representations aggregate consecutive steps into compact tokens to improve efficiency and model local temporal dynamics, but they often discard fine-grained temporal details that are critical for accurate predictions in volatile or complex time series. We propose IPatch, a multi-resolution Transformer architecture that integrates both point-wise and patch-wise tokens, modeling temporal information at multiple resolutions. Experiments on 7 benchmark datasets demonstrate that IPatch consistently improves forecasting accuracy, robustness to noise, and generalization across various prediction horizons compared to single-representation baselines.

cs.LG↗

arXiv2025

A Robust Clustered Federated Learning Approach for Non-IID Data with Quantity Skew

Federated Learning (FL) is a decentralized paradigm that enables a client-server architecture to collaboratively train a global Artificial Intelligence model without sharing raw data, thereby preserving privacy. A key challenge in FL is Non-IID data. Quantity Skew (QS) is a particular problem of Non-IID, where clients hold highly heterogeneous data volumes. Clustered Federated Learning (CFL) is an emergent variant of FL that presents a promising solution to Non-IID problem. It improves models' performance by grouping clients with similar data distributions into clusters. CFL methods generally fall into two operating strategies. In the first strategy, clients select the cluster that minimizes the local training loss. In the second strategy, the server groups clients based on local model similarities. However, most CFL methods lack systematic evaluation under QS but present significant challenges because of it. In this paper, we present two main contributions. The first one is an evaluation of state-of-the-art CFL algorithms under various Non-IID settings, applying multiple QS scenarios to assess their robustness. Our second contribution is a novel iterative CFL algorithm, named CORNFLQS, which proposes an optimal coordination between both operating strategies of CFL. Our approach is robust against the different variations of QS settings. We conducted intensive experiments on six image classification datasets, resulting in 270 Non-IID configurations. The results show that CORNFLQS achieves the highest average ranking in both accuracy and clustering quality, as well as strong robustness to QS perturbations. Overall, our approach outperforms actual CFL algorithms.

cs.LG↗

arXiv2025

Anomaly detection by partitioning of multi-variate time series

In this article, we suggest a novel non-supervised partition based anomaly detection method for anomaly detection in multivariate time series called PARADISE. This methodology creates a partition of the variables of the time series while ensuring that the inter-variable relations remain untouched. This partitioning relies on the clustering of multiple correlation coefficients between variables to identify subsets of variables before executing anomaly detection algorithms locally for each of those subsets. Through multiple experimentations done on both synthetic and real datasets coming from the literature, we show the relevance of our approach with a significant improvement in anomaly detection performance.

cs.LG↗

arXiv2025

XStacking: Explanation-Guided Stacked Ensemble Learning

Ensemble Machine Learning (EML) techniques, especially stacking, have been shown to improve predictive performance by combining multiple base models. However, they are often criticized for their lack of interpretability. In this paper, we introduce XStacking, an effective and inherently explainable framework that addresses this limitation by integrating dynamic feature transformation with model-agnostic Shapley additive explanations. This enables stacked models to retain their predictive accuracy while becoming inherently explainable. We demonstrate the effectiveness of the framework on 29 datasets, achieving improvements in both the predictive effectiveness of the learning space and the interpretability of the resulting models. XStacking offers a practical and scalable solution for responsible ML.

cs.LG↗

arXiv2025

A survey on Clustered Federated Learning: Taxonomy, Analysis and Applications

As Federated Learning (FL) expands, the challenge of non-independent and identically distributed (non-IID) data becomes critical. Clustered Federated Learning (CFL) addresses this by training multiple specialized models, each representing a group of clients with similar data distributions. However, the term ''CFL'' has increasingly been applied to operational strategies unrelated to data heterogeneity, creating significant ambiguity. This survey provides a systematic review of the CFL literature and introduces a principled taxonomy that classifies algorithms into Server-side, Client-side, and Metadata-based approaches. Our analysis reveals a distinct dichotomy: while theoretical research prioritizes privacy-preserving Server/Client-side methods, real-world applications in IoT, Mobility, and Energy overwhelmingly favor Metadata-based efficiency. Furthermore, we explicitly distinguish ''Core CFL'' (grouping clients for non-IID data) from ''Clustered X FL'' (operational variants for system heterogeneity). Finally, we outline lessons learned and future directions to bridge the gap between theoretical privacy and practical efficiency.

stat.ML↗

arXiv2024

Comparative Evaluation of Clustered Federated Learning Methods

Over recent years, Federated Learning (FL) has proven to be one of the most promising methods of distributed learning which preserves data privacy. As the method evolved and was confronted to various real-world scenarios, new challenges have emerged. One such challenge is the presence of highly heterogeneous (often referred as non-IID) data distributions among participants of the FL protocol. A popular solution to this hurdle is Clustered Federated Learning (CFL), which aims to partition clients into groups where the distribution are homogeneous. In the literature, state-of-the-art CFL algorithms are often tested using a few cases of data heterogeneities, without systematically justifying the choices. Further, the taxonomy used for differentiating the different heterogeneity scenarios is not always straightforward. In this paper, we explore the performance of two state-of-theart CFL algorithms with respect to a proposed taxonomy of data heterogeneities in federated learning (FL). We work with three image classification datasets and analyze the resulting clusters against the heterogeneity classes using extrinsic clustering metrics. Our objective is to provide a clearer understanding of the relationship between CFL performances and data heterogeneity scenarios.

stat.ML↗

arXiv2023

DECWA : Density-Based Clustering using Wasserstein Distance

Clustering is a data analysis method for extracting knowledge by discovering groups of data called clusters. Among these methods, state-of-the-art density-based clustering methods have proven to be effective for arbitrary-shaped clusters. Despite their encouraging results, they suffer to find low-density clusters, near clusters with similar densities, and high-dimensional data. Our proposals are a new characterization of clusters and a new clustering algorithm based on spatial density and probabilistic approach. First of all, sub-clusters are built using spatial density represented as probability density function ($p.d.f$) of pairwise distances between points. A method is then proposed to agglomerate similar sub-clusters by using both their density ($p.d.f$) and their spatial distance. The key idea we propose is to use the Wasserstein metric, a powerful tool to measure the distance between $p.d.f$ of sub-clusters. We show that our approach outperforms other state-of-the-art density-based clustering methods on a wide variety of datasets.

cs.LG↗

arXiv2022

Why Should I Choose You? AutoXAI: A Framework for Selecting and Tuning eXplainable AI Solutions

In recent years, a large number of XAI (eXplainable Artificial Intelligence) solutions have been proposed to explain existing ML (Machine Learning) models or to create interpretable ML models. Evaluation measures have recently been proposed and it is now possible to compare these XAI solutions. However, selecting the most relevant XAI solution among all this diversity is still a tedious task, especially when meeting specific needs and constraints. In this paper, we propose AutoXAI, a framework that recommends the best XAI solution and its hyperparameters according to specific XAI evaluation metrics while considering the user's context (dataset, ML model, XAI needs and constraints). It adapts approaches from context-aware recommender systems and strategies of optimization and evaluation from AutoML (Automated Machine Learning). We apply AutoXAI to two use cases, and show that it recommends XAI solutions adapted to the user's needs with the best hyperparameters matching the user's constraints.

cs.LG↗

arXiv2022

Dimensional Data KNN-Based Imputation

Data Warehouses (DWs) are core components of Business Intelligence (BI). Missing data in DWs have a great impact on data analyses. Therefore, missing data need to be completed. Unlike other existing data imputation methods mainly adapted for facts, we propose a new imputation method for dimensions. This method contains two steps: 1) a hierarchical imputation and 2) a k-nearest neighbors (KNN) based imputation. Our solution has the advantage of taking into account the DW structure and dependency constraints. Experimental assessments validate our method in terms of effectiveness and efficiency.

cs.DB↗

arXiv2021

Internal Data Imputation in Data Warehouse Dimensions

Missing values occur commonly in the multidimensional data warehouses. They may generate problems of usefulness of data since the analysis performed on a multidimensional data warehouse is through different dimensions with hierarchies where we can roll up or drill down to the different parameters of analysis. Therefore, it's essential to complete these missing values in order to carry out a better analysis. There are existing data imputation methods which are suitable for numeric data, so they can be applied for fact tables but not for dimension tables. Some other data imputation methods need extra time and effort costs. As consequence, we propose in this article an internal data imputation method for multidimensional data warehouse based on the existing data and considering the intra-dimension and inter-dimension relationships.

cs.DB↗

arXiv2021

An Automatic Schema-Instance Approach for Merging Multidimensional Data Warehouses

Using data warehouses to analyse multidimensional data is a significant task in company decision-making.The data warehouse merging process is composed of two steps: matching multidimensional components and then merging them. Current approaches do not take all the particularities of multidimensional data warehouses into account, e.g., only merging schemata, but not instances; or not exploiting hierarchies nor fact tables. Thus, in this paper, we propose an automatic merging approach for star schema-modeled data warehouses that works at both the schema and instance levels. We also provide algorithms for merging hierarchies, dimensions and facts. Eventually, we implement our merging algorithms and validate them with the use of both synthetic and benchmark datasets.

cs.DB↗

arXiv2020

Automatic Integration Issues of Tabular Data for On-Line Analysis Processing

Companies and individuals produce numerous tabular data. The objective of this position paper is to draw up the challenges posed by the automatic integration of data in the form of tables so that they can be cross-analyzed. We provide a first automatic solution for the integration of such tabular data to allow On-Line Analysis Processing. To fulfil this task, features of tabular data should be analyzed and the challenge of automatic multidimensional schema generation should be addressed. Hence, we propose a typology of tabular data and discuss our idea of an automatic solution.

cs.DB↗

arXiv2010

Personnalisation de Systèmes OLAP Annotés

This paper deals with personalization of annotated OLAP systems. Data constellation is extended to support annotations and user preferences. Annotations reflect the decision-maker experience whereas user preferences enable users to focus on the most interesting data. User preferences allow annotated contextual recommendations helping the decision-maker during his/her multidimensional navigations.

cs.DB↗

arXiv2010

Personnalisation de bases de données multidimensionnelles

This paper deals with decision support systems resting on multidimensional modelling of data. Moreover, we intend to offer a set of concepts and mechanisms for personalized multidimensional database specifications. This personalization consists in associating weights to different components of a multidimensional schema. Personalization specifications are specified through the use of a language based on the principle of Event Condition Action. This personalisation determines multidimensional data display as well as their analyses (with the use of drilling or rotating operations).

cs.DB↗

arXiv2010

Construction graphique d'entrepôts et de magasins de données

Nowadays, decisional systems have became a significant research topic in databases. Data warehouses and data marts are the main elements of such systems. This paper presents our decisional support system. We present graphical interfaces which help the administrator to build data warehouses and data marts. We present a data warehouse building interface based on an object-oriented conceptual model. This model allows the warehouse data historisation at three levels: attribute, class and environment. Also, we present a data mart building interface which allows warehouse data to be reorganised through a multidimensional object-oriented model.

cs.DB↗

arXiv2010

Algèbre OLAP et langage graphique

This article deals with OLAP systems based on multidimensional model. The conceptual model we provide, represents data through a constellation (multi-facts) composed of several multi-hierarchy dimensions. In this model, data are displayed through multidimensional tables. We define a query algebra handling these tables. This user oriented algebra is composed of a closure core of OLAP operators as soon as advanced operators dedicated to complex analysis. Finally, we specify a graphical OLAP language based on this algebra. This language facilitates analyses of decision makers.

cs.DB↗