A.. _tuning optimization:

tuning optimization

serial_gradient_descent

• Gradient Descent

  This finds the minimum of the resonator using the stochastic gradient descent method/algorithm.

  Extra stuff/notes:

  • This is an optimization function

  • you can use it to approach the minimum of any differentiable function

  • it can get stuck in a local minimum instead of finding the global minimum but it is still useful

  • the cost/loss function is the function to be minimized (in this case it would be the function described by the resonator dip?)

  • goal is to minimize the difference between the predicted function and the actual data

  • minimize the sum of squared residuals (SSR)

  • working with gradient descent just means you’re looking for the fastest decrease in the cost/loss function which is determined by the negative gradient