Categories in Deep Learning
Petar discusses the importance of categories in deep learning beyond geometric layers. He highlights the need for models resistant to non-invertible operations, like scaling, in neural networks. Categories offer a new framework to handle complex transformations in computation and nature.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#85 Dr. Petar Veličković (Deepmind) - Categories, Graphs, Reasoning [NEURIPS22 UNPLUGGED]
Related Questions