Autumn School on Recent Advances in Machine Learning

7:30 - 8:30 (1h)

Breakfast

8:45 - 9:45 (1h)

Conformal prediction - from a general introduction towards conditional guarantees (2/3)

Conference room

Aymeric Dieuleveut

9:45 - 10:45 (1h)

Transport based generative modeling and applications to sampling (2/3)

Conference room

Marylou Gabrié

10:45 - 11:15 (30min)

Coffee break

11:15 - 12:15 (1h)

CT - Session 4

Conference room

› Estimating ratios of normalizing constants using stochastic approximation: the SARIS algorithm - Charlotte Baey, Université de Lille 11:15-11:45 (30min)

› Bayesian deep learning, overview and challenges - Julyan Arbel, INRIA 11:45-12:15 (30min)

12:30 - 13:30 (1h)

Lunch

13:30 - 15:30 (2h)

Free time

15:30 - 16:45 (1h15)

CT - Session 5

Conference room

› Equivariant U-Nets for effective image deblurring: design principles and challenges - Jérémy Scanvic, Laboratoire de Physique, ENS de Lyon, Foxstream 15:30-15:55 (25min)

› Conditional Sampling with Score-Based Models via Particle Methods - Stanislas Strasman, Laboratoire de Probabilités, Statistique et Modélisation 15:55-16:20 (25min)

› Selecting informative conformal prediction sets - Ulysse Gazin, Laboratoire de Probabilités, Statistique et Modélisation 16:20-16:45 (25min)

16:45 - 17:15 (30min)

Coffee break

17:15 - 18:05 (50min)

Predictive posterior sampling from non-stationnary Gaussian process priors via Diffusion models with application to climate data.

Conference room

Gabriel Victorino-Cardoso

18:10 - 19:00 (50min)

TBA

Conference room

Aurélien Bellet

19:30 - 20:30 (1h)

Dinner