talks

2025

Diagram comparing joint-embedding and reconstruction objectives

Joint Embedding vs Reconstruction: Which Paradigm to Choose for Pretraining?

Slides Poster

  • · NeurIPS 2025, San Diego
  • · Prescient Design Machine Learning seminar, South San Francisco

2024

Visualization from a talk on dimensionality reduction with optimal transport

Revisiting Dimensionality Reduction with Optimal Transport: Within and Across Spaces

Slides

  • · Genentech, South San Francisco
  • · Memorial Sloan Kettering Cancer Center, New York
Visualization from a talk on distributional reduction

Unifying Dimensionality Reduction and Clustering with Gromov–Wasserstein

Slides

  • · Cambridge University Machine Learning group seminar
  • · CANUM 2024, Île de Ré
  • · LaMME seminar, Université Paris-Saclay
  • · Inria Mind team seminar

2023

Visualization from the SNEkhorn dimensionality-reduction talk

Affinity Matrices with Symmetric Optimal Transport

Slides Poster

  • · NeurIPS 2023, New Orleans
  • · RT MIA thematic day on Dimensionality Reduction, Lyon
  • · PariSanté Campus, Paris
  • · GeMSS 2023, Copenhagen

2022

Visualization from the graph-coupling dimensionality-reduction talk

Graph Coupling for Dimensionality Reduction

Slides Poster

  • · NeurIPS 2022, New Orleans
  • · Statistical Challenges in scRNA-seq Data Analysis workshop, Toulouse
  • · EPIT workshop, CIRM, Marseille
  • · Statlearn workshop, Cargèse
  • · MLSP seminar, ENS Lyon