#neural-networks
3 articles-
AI Recap, Sidebar 2: How Neural Networks Represent the World
From authored ontologies to learned features: how we investigate neural representations, what their geometry can tell us, and where observation gives way to interpretation.
-
AI Recap, Sidebar 1: The Soft Threshold's Bargain
Why sigmoid units made gradient-based training practical, how saturation complicates learning in deep networks, and how later methods address gradient propagation.
-
AI Recap, Part 1: How Neural Networks Survived Their Second Winter
Returning to neural networks after two decades: the training difficulties, competing methods and continuing research that connect backpropagation to AlexNet.