Chapter 9: Classical Optimizers for Variational Quantum Algorithms

This section is already in the book plan, but it has not been written fully yet. The book owner can press Generate section to write this part with the language model connected to TheoryTrace.

Section plan:

Covers gradient-free and gradient-based methods, including COBYLA, Nelder-Mead, SPSA, Adam, parameter-shift gradients, natural gradients, and practical optimizer selection under noisy measurements.