Diffusions now are recurrent: RecMoDiffuse - Recurrent Flow Diffusion for Human Motion Generation
In this talk we explain our recently proposed novel architecture RecMoDiffuse: Recurrent Flow Diffusion, a new recurrent diffusion formulation for temporal modelling. Our method explicitly enforces temporal constraints with the means of normalizing flow models in the diffusion process and thereby extends diffusion to the temporal dimension.
Jan 24, 2025
Statistical ML paradigms
This talk explores the core paradigms of statistical machine learning, highlighting their principles, applications, and impact in creating intelligent systems.
May 10, 2020