GFLO Nets and RL Overlaps
Yoshua discusses the overlaps between GFLO nets and RL, highlighting the focus on sampling trajectory sequences of actions and the importance of exploration. He explains how GFLO nets are particularly interesting for sampling from posterior distributions in a Bayesian sense, and how they are applied in active learning applications.In this clip
From this podcast
The Gradient
Yoshua Bengio: The Past, Present, and Future of Deep Learning
Related Questions