Michèle Sébag
YOU?
Author Swipe
View article: DCILP: A Distributed Approach for Large-Scale Causal Structure Learning
DCILP: A Distributed Approach for Large-Scale Causal Structure Learning Open
Causal learning tackles the computationally demanding task of estimating causal graphs. This paper introduces a new divide-and-conquer approach for causal graph learning, called DCILP. In the divide phase, the Markov blanket MB(Xi) of each…
View article: Asymmetrical Latent Representation for Individual Treatment Effect Modeling
Asymmetrical Latent Representation for Individual Treatment Effect Modeling Open
Conditional Average Treatment Effect (CATE) estimation, at the heart of counterfactual reasoning, is a crucial challenge for causal modeling both theoretically and applicatively, in domains such as healthcare, sociology, or advertising. Bo…
View article: Provably Safeguarding a Classifier from OOD and Adversarial Samples: an Extreme Value Theory Approach
Provably Safeguarding a Classifier from OOD and Adversarial Samples: an Extreme Value Theory Approach Open
This paper introduces a novel method, Sample-efficient Probabilistic Detection using Extreme Value Theory (SPADE), which transforms a classifier into an abstaining classifier, offering provable protection against out-of-distribution and ad…
View article: Variational Multi-Modal Hypergraph Attention Network for Multi-Modal Relation Extraction
Variational Multi-Modal Hypergraph Attention Network for Multi-Modal Relation Extraction Open
Multi-modal relation extraction (MMRE) is a challenging task that seeks to identify relationships between entities with textual and visual attributes. However, existing methods struggle to handle the complexities posed by multiple entity p…
View article: MEGAD: A Memory-Efficient Framework for Large-Scale Attributed Graph Anomaly Detection
MEGAD: A Memory-Efficient Framework for Large-Scale Attributed Graph Anomaly Detection Open
Graph anomaly detection (GAD), with its ability to accurately identify anomalous patterns in graph data, plays a vital role in areas such as network security, social media platforms, and fraud detection. Graph autoencoder-based methods are…
View article: DCILP: A Distributed Approach for Large-Scale Causal Structure Learning
DCILP: A Distributed Approach for Large-Scale Causal Structure Learning Open
Causal learning tackles the computationally demanding task of estimating causal graphs. This paper introduces a new divide-and-conquer approach for causal graph learning, called DCILP. In the divide phase, the Markov blanket MB($X_i$) of e…
View article: Learning Structural Causal Models through Deep Generative Models: Methods, Guarantees, and Challenges
Learning Structural Causal Models through Deep Generative Models: Methods, Guarantees, and Challenges Open
This paper provides a comprehensive review of deep structural causal models (DSCMs), particularly focusing on their ability to answer counterfactual queries using observational data within known causal structures. It delves into the charac…
View article: In Silico Generation of Gene Expression profiles using Diffusion Models
In Silico Generation of Gene Expression profiles using Diffusion Models Open
Motivation RNA-seq data is used for precision medicine (e.g., cancer predictions), which benefits from deep learning approaches to analyze complex gene expression data. However, transcriptomics datasets often have few samples compared to d…
View article: Physics-aware modelling of an accelerated particle cloud
Physics-aware modelling of an accelerated particle cloud Open
International audience
View article: Fairness in job recommendations: estimating, explaining, and reducing gender gaps
Fairness in job recommendations: estimating, explaining, and reducing gender gaps Open
International audience
View article: Toward Job Recommendation for All
Toward Job Recommendation for All Open
This paper presents a job recommendation algorithm designed and validated in the context of the French Public Employment Service. The challenges, owing to the confidential data policy, are related with the extreme sparsity of the interacti…
View article: GAN-based data augmentation for transcriptomics: survey and comparative assessment
GAN-based data augmentation for transcriptomics: survey and comparative assessment Open
Motivation Transcriptomics data are becoming more accessible due to high-throughput and less costly sequencing methods. However, data scarcity prevents exploiting deep learning models’ full predictive power for phenotypes prediction. Artif…
View article: Learning Large Causal Structures from Inverse Covariance Matrix via Sparse Matrix Decomposition
Learning Large Causal Structures from Inverse Covariance Matrix via Sparse Matrix Decomposition Open
Learning causal structures from observational data is a fundamental problem facing important computational challenges when the number of variables is large. In the context of linear structural equation models (SEMs), this paper focuses on …
View article: Learning Large Causal Structures from Inverse Covariance Matrix via\n Sparse Matrix Decomposition
Learning Large Causal Structures from Inverse Covariance Matrix via\n Sparse Matrix Decomposition Open
Learning causal structures from observational data is a fundamental problem\nfacing important computational challenges when the number of variables is\nlarge. In the context of linear structural equation models (SEMs), this paper\nfocuses …
View article: Recommender system in a non-stationary context: recommending job ads in pandemic times
Recommender system in a non-stationary context: recommending job ads in pandemic times Open
International audience
View article: Learning Meta-features for AutoML
Learning Meta-features for AutoML Open
International audience
View article: Frugal Machine Learning
Frugal Machine Learning Open
Machine learning, already at the core of increasingly many systems and applications, is set to become even more ubiquitous with the rapid rise of wearable devices and the Internet of Things. In most machine learning applications, the main …
View article: Designing labor market recommender systems: the importance of job seeker preferences and competition
Designing labor market recommender systems: the importance of job seeker preferences and competition Open
We examine the properties of a recommender algorithm currently under construction at the Public Employment Service (PES) in France, before its implementation in the field. The algorithm associates to each offer-job seeker pair a predicted …
View article: Congestion-Avoiding Job Recommendation with Optimal Transport
Congestion-Avoiding Job Recommendation with Optimal Transport Open
International audience
View article: On the Identifiability of Hierarchical Decision Models
On the Identifiability of Hierarchical Decision Models Open
Interpretability is a desirable property for machine learning and decision models, particularly in the context of safety-critical applications. Another most desirable property of the sought model is to be unique or {\em identifiable} in th…
View article: Towards causal modeling of nutritional outcomes
Towards causal modeling of nutritional outcomes Open
International audience
View article: OcéanIA: AI, Data, and Models for Understanding the Ocean and Climate Change
OcéanIA: AI, Data, and Models for Understanding the Ocean and Climate Change Open
International audience
View article: Boltzman Tuning of Generative Models
Boltzman Tuning of Generative Models Open
The paper focuses on the a posteriori tuning of a generative model in order to favor the generation of good instances in the sense of some external differentiable criterion. The proposed approach, called Boltzmann Tuning of Generative Mode…
View article: Boltzmann Tuning of Generative Models
Boltzmann Tuning of Generative Models Open
The paper focuses on the a posteriori tuning of a generative model in order to favor the generation of good instances in the sense of some external differentiable criterion. The proposed approach, called Boltzmann Tuning of Generative Mode…
View article: Distribution-Based Invariant Deep Networks for Learning Meta-Features
Distribution-Based Invariant Deep Networks for Learning Meta-Features Open
Recent advances in deep learning from probability distributions successfully achieve classification or regression from distribution samples, thus invariant under permutation of the samples. The first contribution of the paper is to extend …
View article: Iterative Learning for Model Reactive Control: Application to Autonomous Multi-agent Control
Iterative Learning for Model Reactive Control: Application to Autonomous Multi-agent Control Open
International audience
View article: Artificial Intelligence, Machine Learning and Modeling for Understanding the Oceans and Climate Change
Artificial Intelligence, Machine Learning and Modeling for Understanding the Oceans and Climate Change Open
International audience