AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Using Sparsity in Deep Learning Networks
Numenta argues that sparsity is key for how information is stored and processed. They also believe it to be one of the most important missing ingredients in modern deep learning. A sparse network is one where all the nurons are not densely connected to every other in the same cortical area. Sparsity will lead to a massively smaller memory footprint, enabling hardware to run more network simultaneously.