Vincent Lemaire

Alexis Bondu, Youssef Achenchabe, Albert Bifet et al.

More and more applications require early decisions, i.e. taken as soon as possible from partially observed data. However, the later a decision is made, the more its accuracy tends to improve, since the description of the problem to hand is enriched over time. Such a compromise between the earliness and the accuracy of decisions has been particularly studied in the field of Early Time Series Classification. This paper introduces a more general problem, called Machine Learning based Early Decision Making (ML-EDM), which consists in optimizing the decision times of models in a wide range of settings where data is collected over time. After defining the ML-EDM problem, ten challenges are identified and proposed to the scientific community to further research in this area. These challenges open important application perspectives, discussed in this paper.

Aurélien Renault, Youssef Achenchabe, Édouard Bertrand et al.

\texttt{ml\_edm} is a Python 3 library, designed for early decision making of any learning tasks involving temporal/sequential data. The package is also modular, providing researchers an easy way to implement their own triggering strategy for classification, regression or any machine learning task. As of now, many Early Classification of Time Series (ECTS) state-of-the-art algorithms, are efficiently implemented in the library leveraging parallel computation. The syntax follows the one introduce in \texttt{scikit-learn}, making estimators and pipelines compatible with \texttt{ml\_edm}. This software is distributed over the BSD-3-Clause license, source code can be found at \url{https://github.com/ML-EDM/ml_edm}.

Colin Troisemaine, Vincent Lemaire, Stéphane Gosselin et al.

Novel Class Discovery (NCD) is a growing field where we are given during training a labeled set of known classes and an unlabeled set of different classes that must be discovered. In recent years, many methods have been proposed to address this problem, and the field has begun to mature. In this paper, we provide a comprehensive survey of the state-of-the-art NCD methods. We start by formally defining the NCD problem and introducing important notions. We then give an overview of the different families of approaches, organized by the way they transfer knowledge from the labeled set to the unlabeled set. We find that they either learn in two stages, by first extracting knowledge from the labeled data only and then applying it to the unlabeled data, or in one stage by conjointly learning on both sets. For each family, we describe their general principle and detail a few representative methods. Then, we briefly introduce some new related tasks inspired by the increasing number of NCD works. We also present some common tools and techniques used in NCD, such as pseudo labeling, self-supervised learning and contrastive learning. Finally, to help readers unfamiliar with the NCD problem differentiate it from other closely related domains, we summarize some of the closest areas of research and discuss their main differences.

Colin Troisemaine, Joachim Flocon-Cholet, Stéphane Gosselin et al.

In Novel Class Discovery (NCD), the goal is to find new classes in an unlabeled set given a labeled set of known but different classes. While NCD has recently gained attention from the community, no framework has yet been proposed for heterogeneous tabular data, despite being a very common representation of data. In this paper, we propose TabularNCD, a new method for discovering novel classes in tabular data. We show a way to extract knowledge from already known classes to guide the discovery process of novel classes in the context of tabular data which contains heterogeneous variables. A part of this process is done by a new method for defining pseudo labels, and we follow recent findings in Multi-Task Learning to optimize a joint objective function. Our method demonstrates that NCD is not only applicable to images but also to heterogeneous tabular data. Extensive experiments are conducted to evaluate our method and demonstrate its effectiveness against 3 competitors on 7 diverse public classification datasets.

Colin Troisemaine, Alexandre Reiffers-Masson, Stéphane Gosselin et al.

The problem of Novel Class Discovery (NCD) consists in extracting knowledge from a labeled set of known classes to accurately partition an unlabeled set of novel classes. While NCD has recently received a lot of attention from the community, it is often solved on computer vision problems and under unrealistic conditions. In particular, the number of novel classes is usually assumed to be known in advance, and their labels are sometimes used to tune hyperparameters. Methods that rely on these assumptions are not applicable in real-world scenarios. In this work, we focus on solving NCD in tabular data when no prior knowledge of the novel classes is available. To this end, we propose to tune the hyperparameters of NCD methods by adapting the $k$-fold cross-validation process and hiding some of the known classes in each fold. Since we have found that methods with too many hyperparameters are likely to overfit these hidden classes, we define a simple deep NCD model. This method is composed of only the essential elements necessary for the NCD problem and performs impressively well under realistic conditions. Furthermore, we find that the latent space of this method can be used to reliably estimate the number of novel classes. Additionally, we adapt two unsupervised clustering algorithms ($k$-means and Spectral Clustering) to leverage the knowledge of the known classes. Extensive experiments are conducted on 7 tabular datasets and demonstrate the effectiveness of the proposed method and hyperparameter tuning process, and show that the NCD problem can be solved without relying on knowledge from the novel classes.

Aurélien Renault, Alexis Bondu, Vincent Lemaire et al.

Time Series Classification (TSC) has received much attention in the past two decades and is still a crucial and challenging problem in data science and knowledge engineering. Indeed, along with the increasing availability of time series data, many TSC algorithms have been suggested by the research community in the literature. Besides state-of-the-art methods based on similarity measures, intervals, shapelets, dictionaries, deep learning methods or hybrid ensemble methods, several tools for extracting unsupervised informative summary statistics, aka features, from time series have been designed in the recent years. Originally designed for descriptive analysis and visualization of time series with informative and interpretable features, very few of these feature engineering tools have been benchmarked for TSC problems and compared with state-of-the-art TSC algorithms in terms of predictive performance. In this article, we aim at filling this gap and propose a simple TSC process to evaluate the potential predictive performance of the feature sets obtained with existing feature engineering tools. Thus, we present an empirical study of 11 feature engineering tools branched with 9 supervised classifiers over 112 time series data sets. The analysis of the results of more than 10000 learning experiments indicate that feature-based methods perform as accurately as current state-of-the-art TSC algorithms, and thus should rightfully be considered further in the TSC literature.

Vincent Lemaire, Fabrice Clérot, Marc Boullé

Variable selection or importance measurement of input variables to a machine learning model has become the focus of much research. It is no longer enough to have a good model, one also must explain its decisions. This is why there are so many intelligibility algorithms available today. Among them, Shapley value estimation algorithms are intelligibility methods based on cooperative game theory. In the case of the naive Bayes classifier, and to our knowledge, there is no ``analytical" formulation of Shapley values. This article proposes an exact analytic expression of Shapley values in the special case of the naive Bayes Classifier. We analytically compare this Shapley proposal, to another frequently used indicator, the Weight of Evidence (WoE) and provide an empirical comparison of our proposal with (i) the WoE and (ii) KernelShap results on real world datasets, discussing similar and dissimilar results. The results show that our Shapley proposal for the naive Bayes classifier provides informative results with low algorithmic complexity so that it can be used on very large datasets with extremely low computation time.

Colin Troisemaine, Joachim Flocon-Cholet, Stéphane Gosselin et al.

Novel Class Discovery (NCD) is the problem of trying to discover novel classes in an unlabeled set, given a labeled set of different but related classes. The majority of NCD methods proposed so far only deal with image data, despite tabular data being among the most widely used type of data in practical applications. To interpret the results of clustering or NCD algorithms, data scientists need to understand the domain- and application-specific attributes of tabular data. This task is difficult and can often only be performed by a domain expert. Therefore, this interface allows a domain expert to easily run state-of-the-art algorithms for NCD in tabular data. With minimal knowledge in data science, interpretable results can be generated.

Arthur Hoarau, Vincent Lemaire

Recent research in machine learning has given rise to a flourishing literature on the quantification and decomposition of model uncertainty. This information can be very useful during interactions with the learner, such as in active learning or adaptive learning, and especially in uncertainty sampling. To allow a simple representation of these total, epistemic (reducible) and aleatoric (irreducible) uncertainties, we offer DEMAU, an open-source educational, exploratory and analytical tool allowing to visualize and explore several types of uncertainty for classification models in machine learning.

Vincent Lemaire, Nathan Le Boudec, Victor Guyomard et al.

There are now many explainable AI methods for understanding the decisions of a machine learning model. Among these are those based on counterfactual reasoning, which involve simulating features changes and observing the impact on the prediction. This article proposes to view this simulation process as a source of creating a certain amount of knowledge that can be stored to be used, later, in different ways. This process is illustrated in the additive model and, more specifically, in the case of the naive Bayes classifier, whose interesting properties for this purpose are shown.

Arthur Hoarau, Vincent Lemaire, Arnaud Martin et al.

Recent studies in active learning, particularly in uncertainty sampling, have focused on the decomposition of model uncertainty into reducible and irreducible uncertainties. In this paper, the aim is to simplify the computational process while eliminating the dependence on observations. Crucially, the inherent uncertainty in the labels is considered, the uncertainty of the oracles. Two strategies are proposed, sampling by Klir uncertainty, which tackles the exploration-exploitation dilemma, and sampling by evidential epistemic uncertainty, which extends the concept of reducible uncertainty within the evidential framework, both using the theory of belief functions. Experimental results in active learning demonstrate that our proposed method can outperform uncertainty sampling.

Youssef Achenchabe, Alexis Bondu, Antoine Cornuéjols et al.

In numerous applications, for instance in predictive maintenance, there is a pression to predict events ahead of time with as much accuracy as possible while not delaying the decision unduly. This translates in the optimization of a trade-off between earliness and accuracy of the decisions, that has been the subject of research for time series of finite length and with a unique label. And this has led to powerful algorithms for Early Classification of Time Series (ECTS). This paper, for the first time, investigates such a trade-off when events of different classes occur in a streaming fashion, with no predefined end. In the Early Classification in Open Time Series problem (ECOTS), the task is to predict events, i.e. their class and time interval, at the moment that optimizes the accuracy vs. earliness trade-off. Interestingly, we find that ECTS algorithms can be sensibly adapted in a principled way to this new problem. We illustrate our methodology by transforming two state-of-the-art ECTS algorithms for the ECOTS scenario. Among the wide variety of applications that this new approach opens up, we develop a predictive maintenance use case that optimizes alarm triggering times, thus demonstrating the power of this new approach.

Pierre Nodet, Vincent Lemaire, Alexis Bondu et al.

Training machine learning models from data with weak supervision and dataset shifts is still challenging. Designing algorithms when these two situations arise has not been explored much, and existing algorithms cannot always handle the most complex distributional shifts. We think the biquality data setup is a suitable framework for designing such algorithms. Biquality Learning assumes that two datasets are available at training time: a trusted dataset sampled from the distribution of interest and the untrusted dataset with dataset shifts and weaknesses of supervision (aka distribution shifts). The trusted and untrusted datasets available at training time make designing algorithms dealing with any distribution shifts possible. We propose two methods, one inspired by the label noise literature and another by the covariate shift literature for biquality learning. We experiment with two novel methods to synthetically introduce concept drift and class-conditional shifts in real-world datasets across many of them. We opened some discussions and assessed that developing biquality learning algorithms robust to distributional changes remains an interesting problem for future research.

Pierre Nodet, Vincent Lemaire, Alexis Bondu et al.

The democratization of Data Mining has been widely successful thanks in part to powerful and easy-to-use Machine Learning libraries. These libraries have been particularly tailored to tackle Supervised Learning. However, strong supervision signals are scarce in practice, and practitioners must resort to weak supervision. In addition to weaknesses of supervision, dataset shifts are another kind of phenomenon that occurs when deploying machine learning models in the real world. That is why Biquality Learning has been proposed as a machine learning framework to design algorithms capable of handling multiple weaknesses of supervision and dataset shifts without assumptions on their nature and level by relying on the availability of a small trusted dataset composed of cleanly labeled and representative samples. Thus we propose biquality-learn: a Python library for Biquality Learning with an intuitive and consistent API to learn machine learning models from biquality data, with well-proven algorithms, accessible and easy to use for everyone, and enabling researchers to experiment in a reproducible way on biquality data.

Colin Troisemaine, Joachim Flocon-Cholet, Stéphane Gosselin et al.

In Novel Class Discovery (NCD), the goal is to find new classes in an unlabeled set given a labeled set of known but different classes. While NCD has recently gained attention from the community, no framework has yet been proposed for heterogeneous tabular data, despite being a very common representation of data. In this paper, we propose TabularNCD, a new method for discovering novel classes in tabular data. We show a way to extract knowledge from already known classes to guide the discovery process of novel classes in the context of tabular data which contains heterogeneous variables. A part of this process is done by a new method for defining pseudo labels, and we follow recent findings in Multi-Task Learning to optimize a joint objective function. Our method demonstrates that NCD is not only applicable to images but also to heterogeneous tabular data.

Aurélien Renault, Alexis Bondu, Antoine Cornuéjols et al.

Early Classification of Time Series (ECTS) addresses decision-making problems in which predictions must be made as early as possible while maintaining high accuracy. Most existing ECTS methods assume that the time-dependent decision costs governing the learning objective are known, fixed, and correctly specified. In practice, however, these costs are often uncertain and may change over time, leading to mismatches between training-time and deployment-time objectives. In this paper, we study ECTS under two practically relevant forms of cost non-stationarity: drift in the balance between misclassification and decision delay costs, and stochastic realizations of decision costs that deviate from the nominal training-time model. To address these challenges, we revisit representative ECTS approaches and adapt them to an online learning setting. Focusing on separable methods, we update only the triggering model during deployment, while keeping the classifier fixed. We propose several online adaptations and baselines, including bandit-based and RL-based approaches, and conduct controlled experiments on synthetic data to systematically evaluate robustness under cost non-stationarity. Our results demonstrate that online learning can effectively improve the robustness of ECTS methods to cost drift, with RL-based strategies exhibiting strong and stable performance across varying cost regimes.

Stanislas Strasman, Gabriel Cardoso, Sylvain Le Corff et al.

Understanding the stability and long-time behavior of generative models is a fundamental problem in modern machine learning. This paper provides quantitative bounds on the sampling error of score-based generative models by leveraging stability and forgetting properties of the Markov chain associated with the reverse-time dynamics. Under weak assumptions, we provide the two structural properties to ensure the propagation of initialization and discretization errors of the backward process: a Lyapunov drift condition and a Doeblin-type minorization condition. A practical consequence is quantitative stability of the sampling procedure, as the reverse diffusion dynamics induces a contraction mechanism along the sampling trajectory. Our results clarify the role of stochastic dynamics in score-based models and provide a principled framework for analyzing propagation of errors in such approaches.

Marc Boullé, Nicolas Voisine, Bruno Guerraz et al.

Khiops is an open source machine learning tool designed for mining large multi-table databases. Khiops is based on a unique Bayesian approach that has attracted academic interest with more than 20 publications on topics such as variable selection, classification, decision trees and co-clustering. It provides a predictive measure of variable importance using discretisation models for numerical data and value clustering for categorical data. The proposed classification/regression model is a naive Bayesian classifier incorporating variable selection and weight learning. In the case of multi-table databases, it provides propositionalisation by automatically constructing aggregates. Khiops is adapted to the analysis of large databases with millions of individuals, tens of thousands of variables and hundreds of millions of records in secondary tables. It is available on many environments, both from a Python library and via a user interface.

Aurélien Renault, Alexis Bondu, Antoine Cornuéjols et al.

In many situations, the measurements of a studied phenomenon are provided sequentially, and the prediction of its class needs to be made as early as possible so as not to incur too high a time penalty, but not too early and risk paying the cost of misclassification. This problem has been particularly studied in the case of time series, and is known as Early Classification of Time Series (ECTS). Although it has been the subject of a growing body of literature, there is still a lack of a systematic, shared evaluation protocol to compare the relative merits of the various existing methods. In this paper, we highlight the two components of an ECTS system: decision and prediction, and focus on the approaches that separate them. This document begins by situating these methods within a principle-based taxonomy. It defines dimensions for organizing their evaluation and then reports the results of a very extensive set of experiments along these dimensions involving nine state-of-the-art ECTS algorithms. In addition, these and other experiments can be carried out using an open-source library in which most of the existing ECTS algorithms have been implemented (see https://github.com/ML-EDM/ml_edm).

Ilies Chibane, Thomas George, Pierre Nodet et al.

Mislabeled data is a pervasive issue that undermines the performance of machine learning systems in real-world applications. An effective approach to mitigate this problem is to detect mislabeled instances and subject them to special treatment, such as filtering or relabeling. Automatic mislabeling detection methods typically rely on training a base machine learning model and then probing it for each instance to obtain a trust score that each provided label is genuine or incorrect. The properties of this base model are thus of paramount importance. In this paper, we investigate the impact of calibrating this model. Our empirical results show that using calibration methods improves the accuracy and robustness of mislabeled instance detection, providing a practical and effective solution for industrial applications.

Thomas George, Pierre Nodet, Alexis Bondu et al.

Mislabeled examples are ubiquitous in real-world machine learning datasets, advocating the development of techniques for automatic detection. We show that most mislabeled detection methods can be viewed as probing trained machine learning models using a few core principles. We formalize a modular framework that encompasses these methods, parameterized by only 4 building blocks, as well as a Python library that demonstrates that these principles can actually be implemented. The focus is on classifier-agnostic concepts, with an emphasis on adapting methods developed for deep learning models to non-deep classifiers for tabular data. We benchmark existing methods on (artificial) Completely At Random (NCAR) as well as (realistic) Not At Random (NNAR) labeling noise from a variety of tasks with imperfect labeling rules. This benchmark provides new insights as well as limitations of existing methods in this setup.

Aurélien Renault, Alexis Bondu, Antoine Cornuéjols et al.

Early Classification of Time Series (ECTS) has been recognized as an important problem in many areas where decisions have to be taken as soon as possible, before the full data availability, while time pressure increases. Numerous ECTS approaches have been proposed, based on different triggering functions, each taking into account various pieces of information related to the incoming time series and/or the output of a classifier. Although their performances have been empirically compared in the literature, no studies have been carried out on the optimality of these triggering functions that involve ``man-tailored'' decision rules. Based on the same information, could there be better triggering functions? This paper presents one way to investigate this question by showing first how to translate ECTS problems into Reinforcement Learning (RL) ones, where the very same information is used in the state space. A thorough comparison of the performance obtained by ``handmade'' approaches and their ``RL-based'' counterparts has been carried out. A second question investigated in this paper is whether a different combination of information, defining the state space in RL systems, can achieve even better performance. Experiments show that the system we describe, called \textsc{Alert}, significantly outperforms its state-of-the-art competitors on a large number of datasets.

Marine Hamon, Vincent Lemaire, Nour Eddine Yassine Nair-Benrekia et al.

To detect anomalies with precision and without prior knowledge in time series, is it better to build a detector from the initial temporal representation, or to compute a new (tabular) representation using an existing automatic variable construction library? In this article, we address this question by conducting an in-depth experimental study for two popular detectors (Isolation Forest and Local Outlier Factor). The obtained results, for 5 different datasets, show that the new representation, computed using the tsfresh library, allows Isolation Forest to significantly improve its performance.

Colin Troisemaine, Vincent Lemaire

This paper proposes a method for the automatic creation of variables (in the case of regression) that complement the information contained in the initial input vector. The method works as a pre-processing step in which the continuous values of the variable to be regressed are discretized into a set of intervals which are then used to define value thresholds. Then classifiers are trained to predict whether the value to be regressed is less than or equal to each of these thresholds. The different outputs of the classifiers are then concatenated in the form of an additional vector of variables that enriches the initial vector of the regression problem. The implemented system can thus be considered as a generic pre-processing tool. We tested the proposed enrichment method with 5 types of regressors and evaluated it in 33 regression datasets. Our experimental results confirm the interest of the approach.

Vincent Lemaire, Nédra Meloulli, Pierre Jaquet

Sepsis remains one of the most complex and heterogeneous syndromes in intensive care, characterized by diverse physiological trajectories and variable responses to treatment. While deep learning models perform well in the early prediction of sepsis, they often lack interpretability and ignore latent patient sub-phenotypes. In this work, we propose a machine learning framework by opening up a new avenue for addressing this issue: a relational approach. Temporal data from electronic medical records (EMRs) are viewed as multivariate patient logs and represented in a relational data schema. Then, a propositionalisation technique (based on classic aggregation/selection functions from the field of relational data) is applied to construct interpretable features to "flatten" the data. Finally, the flattened data is classified using a selective naive Bayesian classifier. Experimental validation demonstrates the relevance of the suggested approach as well as its extreme interpretability. The interpretation is fourfold: univariate, global, local, and counterfactual.

Vincent Lemaire, Gilles Pagès, Christian Yeo

We introduce a new class of neural networks designed to be convex functions of their inputs, leveraging the principle that any convex function can be represented as the supremum of the affine functions it dominates. These neural networks, inherently convex with respect to their inputs, are particularly well-suited for approximating the prices of options with convex payoffs. We detail the architecture of this, and establish theoretical convergence bounds that validate its approximation capabilities. We also introduce a \emph{scrambling} phase to improve the training of these networks. Finally, we demonstrate numerically the effectiveness of these networks in estimating prices for three types of options with convex payoffs: Basket, Bermudan, and Swing options.

Colin Troisemaine, Vincent Lemaire

This paper proposes a method for the automatic creation of variables (in the case of regression) that complement the information contained in the initial input vector. The method works as a pre-processing step in which the continuous values of the variable to be regressed are discretized into a set of intervals which are then used to define value thresholds. Then classifiers are trained to predict whether the value to be regressed is less than or equal to each of these thresholds. The different outputs of the classifiers are then concatenated in the form of an additional vector of variables that enriches the initial vector of the regression problem. The implemented system can thus be considered as a generic pre-processing tool. We tested the proposed enrichment method with 5 types of regressors and evaluated it in 33 regression datasets. Our experimental results confirm the interest of the approach.

Youssef Achenchabe, Alexis Bondu, Antoine Cornuéjols et al.

Many approaches have been proposed for early classification of time series in light of itssignificance in a wide range of applications including healthcare, transportation and fi-nance. Until now, the early classification problem has been dealt with by considering onlyirrevocable decisions. This paper introduces a new problem calledearly and revocabletimeseries classification, where the decision maker can revoke its earlier decisions based on thenew available measurements. In order to formalize and tackle this problem, we propose anew cost-based framework and derive two new approaches from it. The first approach doesnot consider explicitly the cost of changing decision, while the second one does. Exten-sive experiments are conducted to evaluate these approaches on a large benchmark of realdatasets. The empirical results obtained convincingly show (i) that the ability of revok-ing decisions significantly improves performance over the irrevocable regime, and (ii) thattaking into account the cost of changing decision brings even better results in general.Keywords:revocable decisions, cost estimation, online decision making

Pierre Nodet, Vincent Lemaire, Alexis Bondu et al.

In this paper we show that the combination of a Contrastive representation with a label noise-robust classification head requires fine-tuning the representation in order to achieve state-of-the-art performances. Since fine-tuned representations are shown to outperform frozen ones, one can conclude that noise-robust classification heads are indeed able to promote meaningful representations if provided with a suitable starting point. Experiments are conducted to draw a comprehensive picture of performances by featuring six methods and nine noise instances of three different kinds (none, symmetric, and asymmetric). In presence of noise the experiments show that fine tuning of Contrastive representation allows the six methods to achieve better results than end-to-end learning and represent a new reference compare to the recent state of art. Results are also remarkable stable versus the noise level.

Youssef Achenchabe, Alexis Bondu, Antoine Cornuéjols et al.

Many approaches have been proposed for early classification of time series in light of its significance in a wide range of applications including healthcare, transportation and finance. However, recently a preprint saved on Arxiv claim that all research done for almost 20 years now on the Early Classification of Time Series is useless, or, at the very least, ill-oriented because severely lacking a strong ground. In this paper, we answer in detail the main issues and misunderstandings raised by the authors of the preprint, and propose directions to further expand the fields of application of early classification of time series.

Dominique Gay, Alexis Bondu, Vincent Lemaire et al.

Supervised learning of time series data has been extensively studied for the case of a categorical target variable. In some application domains, e.g., energy, environment and health monitoring, it occurs that the target variable is numerical and the problem is known as time series extrinsic regression (TSER). In the literature, some well-known time series classifiers have been extended for TSER problems. As first benchmarking studies have focused on predictive performance, very little attention has been given to interpretability. To fill this gap, in this paper, we suggest an extension of a Bayesian method for robust and interpretable feature construction and selection in the context of TSER. Our approach exploits a relational way to tackle with TSER: (i), we build various and simple representations of the time series which are stored in a relational data scheme, then, (ii), a propositionalisation technique (based on classical aggregation / selection functions from the relational data field) is applied to build interpretable features from secondary tables to "flatten" the data; and (iii), the constructed features are filtered out through a Bayesian Maximum A Posteriori approach. The resulting transformed data can be processed with various existing regressors. Experimental validation on various benchmark data sets demonstrates the benefits of the suggested approach.

Vincent Lemaire, Oumaima Alaoui Ismaili, Antoine Cornuéjols et al.

Supervised classification can be effective for prediction but sometimes weak on interpretability or explainability (XAI). Clustering, on the other hand, tends to isolate categories or profiles that can be meaningful but there is no guarantee that they are useful for labels prediction. Predictive clustering seeks to obtain the best of the two worlds. Starting from labeled data, it looks for clusters that are as pure as possible with regards to the class labels. One technique consists in tweaking a clustering algorithm so that data points sharing the same label tend to aggregate together. With distance-based algorithms, such as k-means, a solution is to modify the distance used by the algorithm so that it incorporates information about the labels of the data points. In this paper, we propose another method which relies on a change of representation guided by class densities and then carries out clustering in this new representation space. We present two new algorithms using this technique and show on a variety of data sets that they are competitive for prediction performance with pure supervised classifiers while offering interpretability of the clusters discovered.

Louis Desreumaux, Vincent Lemaire

Active learning aims to reduce annotation cost by predicting which samples are useful for a human expert to label. Although this field is quite old, several important challenges to using active learning in real-world settings still remain unsolved. In particular, most selection strategies are hand-designed, and it has become clear that there is no best active learning strategy that consistently outperforms all others in all applications. This has motivated research into meta-learning algorithms for "learning how to actively learn". In this paper, we compare this kind of approach with the association of a Random Forest with the margin sampling strategy, reported in recent comparative studies as a very competitive heuristic. To this end, we present the results of a benchmark performed on 20 datasets that compares a strategy learned using a recent meta-learning algorithm with margin sampling. We also present some lessons learned and open future perspectives.

Pierre Nodet, Vincent Lemaire, Alexis Bondu et al.

The field of Weakly Supervised Learning (WSL) has recently seen a surge of popularity, with numerous papers addressing different types of "supervision deficiencies". In WSL use cases, a variety of situations exists where the collected "information" is imperfect. The paradigm of WSL attempts to list and cover these problems with associated solutions. In this paper, we review the research progress on WSL with the aim to make it as a brief introduction to this field. We present the three axis of WSL cube and an overview of most of all the elements of their facets. We propose three measurable quantities that acts as coordinates in the previously defined cube namely: Quality, Adaptability and Quantity of information. Thus we suggest that Biquality Learning framework can be defined as a plan of the WSL cube and propose to re-discover previously unrelated patches in WSL literature as a unified Biquality Learning literature.

Hugo Le Baher, Vincent Lemaire, Romain Trinquart

In some industrial applications such as fraud detection, the performance of common supervision techniques may be affected by the poor quality of the available labels : in actual operational use-cases, these labels may be weak in quantity, quality or trustworthiness. We propose a benchmark to evaluate the natural robustness of different algorithms taken from various paradigms on artificially corrupted datasets, with a focus on noisy labels. This paper studies the intrinsic robustness of some leading classifiers. The algorithms under scrutiny include SVM, logistic regression, random forests, XGBoost, Khiops. Furthermore, building on results from recent literature, the study is supplemented with an investigation into the opportunity to enhance some algorithms with symmetric loss functions.

Pierre Nodet, Vincent Lemaire, Alexis Bondu et al.

The field of Weakly Supervised Learning (WSL) has recently seen a surge of popularity, with numerous papers addressing different types of "supervision deficiencies", namely: poor quality, non adaptability, and insufficient quantity of labels. Regarding quality, label noise can be of different types, including completely-at-random, at-random or even not-at-random. All these kinds of label noise are addressed separately in the literature, leading to highly specialized approaches. This paper proposes an original, encompassing, view of Weakly Supervised Learning, which results in the design of generic approaches capable of dealing with any kind of label noise. For this purpose, an alternative setting called "Biquality data" is used. It assumes that a small trusted dataset of correctly labeled examples is available, in addition to an untrusted dataset of noisy examples. In this paper, we propose a new reweigthing scheme capable of identifying noncorrupted examples in the untrusted dataset. This allows one to learn classifiers using both datasets. Extensive experiments that simulate several types of label noise and that vary the quality and quantity of untrusted examples, demonstrate that the proposed approach outperforms baselines and state-of-the-art approaches.

Dominique Gay, Vincent Lemaire

Since the introduction and the public availability of the \textsc{ucr} time series benchmark data sets, numerous Time Series Classification (TSC) methods has been designed, evaluated and compared to each others. We suggest a critical view of TSC performance evaluation protocols put in place in recent TSC literature. The main goal of this `position' paper is to stimulate discussion and reflexion about performance evaluation in TSC literature.

Colin Leverger, Vincent Lemaire, Simon Malinowski et al.

In the context of capacity planning, forecasting the evolution of informatics servers usage enables companies to better manage their computational resources. We address this problem by collecting key indicator time series and propose to forecast their evolution a day-ahead. Our method assumes that data is structured by a daily seasonality, but also that there is typical evolution of indicators within a day. Then, it uses the combination of a clustering algorithm and Markov Models to produce day-ahead forecasts. Our experiments on real datasets show that the data satisfies our assumption and that, in the case study, our method outperforms classical approaches (AR, Holt-Winters).

Patrick Luciano, Ismail Rebai, Vincent Lemaire

Customer Relationship Management (CRM) is a key element of modern marketing strategies. One of the most practical way to build useful knowledge on customers in a CRM system to produce scores to forecast churn behavior, propensity to subscribe to a new service... In AMEA zone (Asia, Middle East and Africa zone), the context of fierce competition may represent a higher percentage, and particularly in B2B market. But by contrast, to our knowledge, no scientific papers were dedicated and published to detail the way to improve loyalty in B2B Telco market. If we can assume at low segments similarities between B2B and B2C, some research is required in order to model B2B user behavior versus B2C behavior. This problematic stands actual as "Bring Your own Device" (BYOD) becomes more and more trendy. Answering business requirements, our team applied some B2C predictive tools with adapting them to B2B in an AMEA country Orange affiliate.