Échapper à la Malédiction de la Dimensionnalité Grâce aux Interpolations de Lois
Centre de recherches mathématiques - CRM via YouTube
Learn Python with Generative AI - Self Paced Online
Google AI Professional Certificate - Learn AI Skills That Get You Hired
Overview
Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Discover key insights in this French-language CRM-SSC Prize lecture that explores methods to overcome the "curse of dimensionality" through law interpolation techniques. Learn how law interpolation methods can be applied even when dimensionality is undefined within the abstract framework of Lebesgue integration, with practical applications in Bayesian data analysis. Gain an undergraduate-level introduction to integral approximation methods based on law interpolations, followed by recent findings comparing two families of law interpolation: MCMC approaches (parallel tempering, simulated tempering) and AIS/SMC approaches (Annealed Importance Sampling, Sequential Monte Carlo).
Syllabus
Alexandre Cöté: Échapper à la malédiction de la dimensionnalité grâce aux interpolations de lois
Taught by
Centre de recherches mathématiques - CRM