gen-briefly-explain-gradient-descent-and-one-common-misconceptio-7a82

Briefly explain gradient descent and one common misconception about it.

# Gradient Descent **Gradient descent** is an optimization algorithm that finds the minimum of a function by iteratively moving in the direction of steepest descent. It calculates the gradient (slope) at the current position and takes a step in the opposite direction, repeating until convergence. The update rule is: **θ = θ - α∇f(θ)** where θ is the parameter, α is the learning rate, and ∇f(θ) is the gradient. ## Common Misconception **"Gradient descent always finds the global minimum"** This is false. Gradient descent only guarantees finding a *local* minimum. For non-convex functions (like neural network loss landscapes), it can get stuck in local minima or saddle points. It only guarantees finding the global minimum for convex functions. This is why initialization, learning rate, and advanced variants (momentum, Adam, etc.) matter significantly in practice.