Dropout as Shared-Parameter Bagging
March 7, 2026
A concrete way to understand dropout in deep learning: sampled subnetworks with shared weights, standing in for a much more expensive ensemble.
Deep Learning
Notes on the mathematics, feedforward networks, and regularization mechanisms in Goodfellow, Bengio, and Courville's Deep Learning.
3 posts
Depth Changes Scaling, Not Just Capacity
February 15, 2026
Chapter 6 separates representational capacity, parameter efficiency, and learnability—and shows why depth changes scaling, not just expressiveness.
Math Foundations for Deep Learning
February 7, 2026
Part I of Goodfellow, Bengio, and Courville works best as shared vocabulary for later chapters, with proof details still worth supplementing.