IDNLearn.com: Where your questions are met with thoughtful and precise answers. Discover the reliable solutions you need with help from our comprehensive and accurate Q&A platform.

Which of the following statements is true about the cost function used for training a neural network through the
gradient descent method?
- The cost function is never a differentiable function of the parameters of the network.
- The cost function is always a linear function of the parameters of the network.
- The cost function can have multiple local minima.
- The cost function is always a convex function of the parameters of the network