Chain Rule
Domain: Math & Logic for AI
🎯 Motivation
Lets us compute derivatives of composite functions. Essential for backpropagation in neural networks.
📋 Concrete Example
If y = f(g(x)), then dy/dx = f-prime(g(x)) times g-prime(x). This is how backprop propagates gradients layer by layer.
🔗 Analogy
Like tracking a rumor through a chain of friends - each person changes it a bit, and you multiply all the changes.