Deep Learning Architectures
Tim discusses the theoretical links between deep learning model architectures on sets, emphasizing the importance of designing algorithms that are invariant to semantically equivalent transformations while maintaining expressivity. Fabian's insights on Genossi pooling and approximate permutation invariance shed light on achieving permutation invariance in computationally tractable ways.In this clip
From this podcast

Machine Learning Street Talk (MLST)
#60 Geometric Deep Learning Blueprint (Special Edition)
Related Questions