Prati
Théo Uscidda
Théo Uscidda
CREST - ENSAE
Potvrđena adresa e-pošte na ensae.fr - Početna stranica
Naslov
Citirano
Citirano
Godina
The monge gap: A regularizer to learn all transport maps
T Uscidda, M Cuturi
International Conference on Machine Learning, 34709-34733, 2023
302023
Unbalancedness in Neural Monge Maps Improves Unpaired Domain Translation
L Eyring*, D Klein, T Uscidda*, G Palla, N Kilbertus, Z Akata, F Theis
The Twelveth International Conference on Learning Representations, 2023
82023
Mirror and Preconditioned Gradient Descent in Wasserstein Space
C Bonet, T Uscidda, A David, PC Aubin-Frankowski, A Korba
The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024
62024
GENOT: Entropic (Gromov) Wasserstein Flow Matching with Applications to Single-Cell Genomics
D Klein*, T Uscidda*, FJ Theis
The Thirty-eighth Annual Conference on Neural Information Processing Systems, 0
6*
On the potential of Optimal Transport in geospatial data science
N Wiedemann, T Uscidda, M Raubal
The Twelveth International Conference on Learning Representations Workshop …, 2024
2024
Disentangled Representation Learning with the Gromov-Monge Gap
T Uscidda*, L Eyring*, K Roth, F Theis, Z Akata, M Cuturi
The Thirteenth International Conference on Learning Representations, 2024
2024
Sustav trenutno ne može provesti ovu radnju. Pokušajte ponovo kasnije.
Članci 1–6