PhD candidate, Machine Learning
SerVal group, SnT, University of Luxembourg
Luxembourg
I'm a PhD candidate in Machine Learning at the SnT of the University of Luxembourg, supervised by Prof. Yves Le Traon and Dr. Sylvain Kubler. My research focuses on synthetic data generation for time series: producing, selecting, and evaluating augmented samples that help downstream models when real data is scarce, imbalanced, or drawn from a different distribution.
In BALANCER (IEEE Access 2024), we trained a model to pick the right augmentation method for each imbalanced dataset. DADA (PHM 2024) uses a Wasserstein CycleGAN to translate labeled source-domain samples into synthetic target-domain samples, allowing a remaining-useful-life predictor to be trained for a target domain that has no labels of its own. ASCENSION (under review at TMLR) expands class regions inside a VAE latent space so a classifier stays confident with little training data. Reach out if any of this is relevant to your work.
Augmentation in the VAE latent space via class-conditional contrastive expansion with an α-scaling control. Evaluated across the 102 UCR datasets; the framework consistently improves downstream classification accuracy over standard augmentation baselines.
A multi-agent system that automates the classification and tagging of predictive-maintenance datasets, lowering the curation effort for downstream learning pipelines.
A Wasserstein CycleGAN with gradient penalty (W-CycleGAN-GP) generates synthetic samples in an unlabeled target domain from labeled source-domain samples. Combined with adversarial and correlation-alignment domain-adaptation losses, the augmented data improves remaining-useful-life prediction on NASA's C-MAPSS turbofan benchmark.
BALANCER is a machine-learning framework that recommends the most effective augmentation technique for an imbalanced time-series classification task. Validated on an empirical study of 720 datasets and explained with SHAP-based interpretability.
If you're a student interested in time series, augmentation, or synthetic data generation, feel free to reach out.