Exercise · ZW-CONVERGENCE-WITH-LIPSCHITZ-GRADIENT-005 · ★☆☆ Basic

If \( \nabla f \) is Lipschitz, what does convexity of \( f \) add for the descent sequence?

Choose one option.

Show answer
  • A. It allows bounding \( f(x_k)-f^* \) at rate \( O(1/k) \)

Explanation

Convexity allows using the gradient inequality to obtain the \( O(1/k) \) bound. The other options contradict the basic results of the method.

Study Continuous optimization in ZWMath →

Review the concept: Convergence with Lipschitz gradient

Result guaranteeing that gradient descent with a suitable step size converges for differentiable functions whose gradient \( \nabla f \) is Lipschitz continuous with constant \( L \).

\[f(x_k)-f^* \le \frac{L\|x_0-x^*\|^2}{2k}\]

More practice problems: convergence with lipschitz gradient

← ZW-CONVERGENCE-WITH-LIPSCHITZ-GRADIENT-004 ZW-CONVERGENCE-WITH-LIPSCHITZ-GRADIENT-006 →