Scholar
Hugues Van Assel
Google Scholar ID: 9Lf9wq8AAAAJ
Genentech
Citations & Impact
Citations
53
H-index
4
i10-index
3
Publications
14
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Developed TorchDR: a modular, GPU-friendly toolbox for dimensionality reduction offering a unified interface for state-of-the-art methods.
- Developed stable-pretraining: a PyTorch library for foundation model pretraining with real-time monitoring.
- Published 'Joint Embedding vs Reconstruction: Provable Benefits of Latent Space Prediction for Self-Supervised Learning' at NeurIPS 2025 (Spotlight).
- Published 'Distributional Reduction: Unifying Dimensionality Reduction and Clustering with Gromov-Wasserstein' in TMLR 2024.
- Published 'SNEkhorn: Dimension Reduction with Symmetric Entropic Affinities' at NeurIPS 2023.
- Published 'A Probabilistic Graph Coupling View of Dimension Reduction' at NeurIPS 2022.
Background
- Currently a Postdoctoral Fellow at Genentech, working with Aviv Regev and Tommaso Biancalani.
- Interested in how machines learn rich and reliable representations of complex data.
- Research focuses on representation learning, self-supervised and multi-modal methods, optimal transport, and dimensionality reduction.
- Develops computational approaches to uncover data structure, motivated by challenges in the life sciences.
- Enjoys building and sharing open-source tools.
Co-authors