Latent Predictive Models
Ishan discusses the challenges of future prediction in vision tasks due to uncertainty, highlighting the role of latent variable models in capturing this uncertainty. He delves into the trade-off between the size of latent variables and the model's predictive power, envisioning future models that can dynamically adjust latent variable sizes based on input uncertainty levels.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#55 Self-Supervised Vision Models (Dr. Ishan Misra - FAIR).
Related Questions