Publications
A General Framework for Joint Multi-State Models
Félix Laplante, Christophe Ambroise
arXiv preprint, October 2025
This paper proposes a unified framework for jointly modeling longitudinal biomarkers and multi-state event processes on directed graphs, accommodating both Markovian and semi-Markovian structures with recurrent cycles and terminal absorptions. Complete likelihood, model selection criteria, and scalable inference procedures based on stochastic gradient descent enable large-scale applications. Nonlinear longitudinal submodels are integrated with shared latent structures, and a dynamic prediction framework provides individualized state-transition probabilities and personalized risk assessments. The approach is validated through simulations and application to the PAQUID cohort, showing accurate parameter recovery and individualized predictions.
@misc{laplante2025joint,
title = {A General Framework for Joint Multi-State Models},
author = {Laplante, F{\'e}lix and Ambroise, Christophe},
year = {2025},
eprint = {2510.07128},
archiveprefix = {arXiv},
primaryclass = {stat.ME},
url = {https://arxiv.org/abs/2510.07128}
}
Spectral Bridges: Scalable Spectral Clustering Based on Vector Quantization
Félix Laplante, Christophe Ambroise
Computo, December 2024
In this paper, Spectral Bridges, a novel clustering algorithm, is introduced. This algorithm builds upon the traditional k-means and spectral clustering frameworks by subdividing data into small Voronoï regions, which are subsequently merged according to a connectivity measure. Drawing inspiration from Support Vector Machine’s margin concept, a non-parametric clustering approach is proposed, building an affinity margin between each pair of Voronoï regions. This approach delineates intricate, non-convex cluster structures and is robust to hyperparameter choice. The numerical experiments underscore Spectral Bridges as a fast, robust, and versatile tool for clustering tasks spanning diverse domains. Its efficacy extends to large-scale scenarios encompassing both real-world and synthetic datasets.
@article{laplante2024spectral,
title = {Spectral Bridges: Scalable Spectral Clustering Based on Vector Quantization},
author = {Laplante, F{\'e}lix and Ambroise, Christophe},
journal = {Computo},
year = {2024},
doi = {10.57750/1gr8-bk61},
issn = {2824-7795},
publisher = {French Statistical Society}
}