Unraveling Contrastive Learning
Ishan delves into the complexities of contrastive learning, highlighting the significance of the momentum encoder in preventing model collapse. He explores the evolution from siamese networks to asymmetric optimization, shedding light on the delicate balance between similarity and dissimilarity assumptions in training models.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#55 Self-Supervised Vision Models (Dr. Ishan Misra - FAIR).
Related Questions