Discreteness, generalization, and GFlowNets

Xu JI
Mila, Québec
https://scholar.google.com/citations?user=HQzOgKYAAAAJ&hl=zh-CN

Date(s) : 28/01/2022   iCal
15 h 30 min - 16 h 30 min

One of the hallmarks of human cognition is the ability to manipulate discrete representations of data. In this talk I will discuss the connection between discretization and generalization, and an idea recently proposed by our group, GFlowNets, for learning to generate discrete and continuous data with probability proportional to their reward.

Bio:
Xu is postdoc at Mila under Yoshua Bengio, formerly PhD student (2021) at Oxford University.

Catégories



Retour en haut 

Secured By miniOrange