Lesson 1.5: Gradient Descent Training

Gradient descent is an iterative optimization algorithm that minimizes the loss function by adjusting weights in the direction of steepest descent.

How it works:

• Calculates gradient (∂Loss/∂W) showing how much each weight contributes to error
• Updates weights using: W_new = W_old - η*(∂Loss/∂W) (η = learning rate)
• Repeats until convergence