How to represent part-whole hierarchies in a neural net
I will present a single idea about representation which allows several recent advances in neural networks to be combined into an imaginary system called GLOM. GLOM answers the question: How can a neural network with a fixed architecture parse an image into a part-whole hierarchy which has a different structure for each image? The idea is simply to use islands of identical vectors to represent the nodes in the parse tree. The talk will discuss the many ramifications of this idea. If GLOM can be made to work, it should significantly improve the interpretability of the representations produced by neural nets when applied to vision or language.