cv

AI Researcher

Sigma Nova

Conducting research primarily on the fine-tuning and adaptation of large pre-trained foundation models.

• Foundation Models
• Model Fine-tuning
• Model Adaptation

Applied AI Researcher

GetVocal

R&D of conversational agents, focusing on automatic speech recognition, large language models and speech synthesis.

• Conversational AI
• Speech Recognition
• Large Language Models
• Speech Synthesis

PhD Student

Commissariat pour l'Énergie Atomique

Studied the contributions of Optimal Transport for Machine Learning, Transfer Learning and Domain Adaptation.

• Optimal Transport
• Transfer Learning
• Domain Adaptation
• Machine Learning

Scientific Researcher in Data Mining

Dell Lead

Mining textual data, building visualizations and exploring strategies for machine reading comprehension.

• Data Mining
• Text Analysis
• Machine Reading Comprehension
• Data Visualization

Research Intern in Artificial Intelligence

Commissariat pour l'Énergie Atomique

Implementing Optimal Transport for Domain Adaptation methods for transfer learning on bacteria cell stain classification.

• Optimal Transport
• Domain Adaptation
• Transfer Learning
• Computer Vision

Research Intern in Deep Learning for Image Restoration

Institut d'Électronique et Télécommunications de Rennes

Implementing software for comparing image restoration methods, with focus on deep neural network methods. Open-source solution hosted on Github.

• Deep Learning
• Image Restoration
• Open Source Software
• Neural Networks

Research Intern in Pattern Recognition

Universidade Federal do Ceará

Working on Computer Vision for Detection and Identification of Cervix Diseases (SISCOLO) project, studying and implementing classifiers for identifying cancer on cervix cells.

• Computer Vision
• Pattern Recognition
• Medical Imaging
• Cancer Detection

PhD

Université Paris-Saclay

Data Science

• Multi-Source Domain Adaptation through Wasserstein Barycenters

Master's Degree

INSA de Rennes

Electronic Engineering and Industrial Informatics

• Optimal Transport for Domain Adaptation with Applications to Bacteria Stain Classification

Bachelor's Degree

Universidade Federal do Ceará

Computer Engineering

• Cross-Domain Fault Diagnosis through Optimal Transport

BRAFITEC/France program

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Scholarship for international exchange program between Brazil and France.

Optimal Transport for Domain Adaptation through Gaussian Mixture Models

Transactions on Machine Learning Research

Machine learning systems operate under the assumption that training and test data are sampled from a fixed probability distribution. However, this assumptions is rarely verified in practice, as the conditions upon which data was acquired are likely to change. In this context, the adaptation of the unsupervised domain requires minimal access to the data of the new conditions for learning models robust to changes in the data distribution. Optimal transport is a theoretically grounded tool for analyzing changes in distribution, especially as it allows the mapping between domains. However, these methods are usually computationally expensive as their complexity scales cubically with the number of samples. In this work, we explore optimal transport between Gaussian Mixture Models (GMMs), which is conveniently written in terms of the components of source and target GMMs. We experiment with 9 benchmarks, with a total of 85 adaptation tasks, showing that our methods are more efficient than previous shallow domain adaptation methods, and they scale well with number of samples n and dimensions d.

Lighter, better, faster multi-source domain adaptation with gaussian mixture models and optimal transport

Joint European Conference on Machine Learning and Knowledge Discovery in Databases

In this paper, we tackle Multi-Source Domain Adaptation (MSDA), a task in transfer learning where one adapts multiple heterogeneous, labeled source probability measures towards a different, unlabeled target measure. We propose a novel framework for MSDA, based on Optimal Transport (OT) and Gaussian Mixture Models (GMMs). Our framework has two key advantages. First, OT between GMMs can be solved efficiently via linear programming. Second, it provides a convenient model for supervised learning, especially classification, as components in the GMM can be associated with existing classes. Based on the GMM-OT problem, we propose a novel technique for calculating barycenters of GMMs. Based on this novel algorithm, we propose two new strategies for MSDA: GMM-Wasserstein Barycenter Transport (WBT) and GMM-Dataset Dictionary Learning (DaDiL). We empirically evaluate our proposed methods on four benchmarks in image classification and fault diagnosis, showing that we improve over the prior art while being faster and involving fewer parameters. Our code is publicly available at https://github.com/eddardd/gmm_msda

Recent advances in optimal transport for machine learning

IEEE Transactions on Pattern Analysis and Machine Intelligence

Recently, Optimal Transport has been proposed as a probabilistic framework in Machine Learning for comparing and manipulating probability distributions. This is rooted in its rich history and theory, and has offered new solutions to different problems in machine learning, such as generative modeling and transfer learning. In this survey we explore contributions of Optimal Transport for Machine Learning over the period 2012 -- 2023, focusing on four sub-fields of Machine Learning: supervised, unsupervised, transfer and reinforcement learning. We further highlight the recent development in computational Optimal Transport and its extensions, such as partial, unbalanced, Gromov and Neural Optimal Transport, and its interplay with Machine Learning practice.

Multi-Source Domain Adaptation Through Dataset Dictionary Learning in Wasserstein Space

26th European Conference on Artificial Intelligence

This paper seeks to solve Multi-Source Domain Adaptation (MSDA), which aims to mitigate data distribution shifts when transferring knowledge from multiple labeled source domains to an unlabeled target domain. We propose a novel MSDA framework based on dictionary learning and optimal transport. We interpret each domain in MSDA as an empirical distribution. As such, we express each domain as a Wasserstein barycenter of dictionary atoms, which are empirical distributions. We propose a novel algorithm, DaDiL, for learning via mini-batches: (i) atom distributions; (ii) a matrix of barycentric coordinates. Based on our dictionary, we propose two novel methods for MSDA: DaDil-R, based on the reconstruction of labeled samples in the target domain, and DaDiL-E, based on the ensembling of classifiers learned on atom distributions. We evaluate our methods in 3 benchmarks: Caltech-Office, Office 31, and CRWU, where we improved previous state-of-the-art by 3.15%, 2.29%, and 7.71% in classification performance. Finally, we show that interpolations in the Wasserstein hull of learned atoms provide data that can generalize to the target domain.

Wasserstein Barycenter for Multi-Source Domain Adaptation

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

Multi-source domain adaptation is a key technique that allows a model to be trained on data coming from various probability distribution. To overcome the challenges posed by this learning scenario, we propose a method for constructing an intermediate domain between sources and target domain, the Wasserstein Barycenter Transport (WBT). This method relies on the barycenter on Wasserstein spaces for aggregating the source probability distributions. Once the sources have been aggregated, they are transported to the target domain using standard Optimal Transport for Domain Adaptation framework. Additionally, we revisit previous single-source domain adaptation tasks in the context of multi-source scenario. In particular, we apply our algorithm to object and face recognition datasets. Moreover, to diversify the range of applications, we also examine the tasks of music genre recognition and music-speech discrimination. The experiments show that our method has similar performance with the existing state-of-the-art.

OpenDenoising

An extensible benchmark for building comparative studies of image denoisers with focus on deep neural network methods.

• Open Source
• Image Restoration
• Deep Learning
• Benchmarking