train the Gradient Descent algorithm for finding the predicted outcome. The predicted outcome is Record Sales. Advertising Budget is the independent variable. Answer the following questions.
a. What is the slope obtained from Gradient Descent?
b. What is the constant obtained from Gradient Descent?
c. At what learning rate convergence was obtained?
d. In how many iterations were the values of slope and constant were obtained with an error of less than 5%?
e. Write your mathematical model with error.
a. Gradient descent is a set of functions that 1) automatically determine the slope in all directions at any given position and 2) adjust the equation's parameters to travel in the direction of the negative slope. This eventually reduces you to a bare minimum.
b. Gradient Descent is an optimization approach for locating a differentiable function's local minimum. Gradient descent is essentially a method for determining the values of a function's parameters (coefficients) that minimize a cost function as much as feasible. To fully grasp this notion, you must first comprehend gradients.
c. The learning rate is a hyper-parameter that governs how much we alter the weights of our network about the loss gradient. Furthermore, our model's learning rate influences how soon it may converge to a local minimum (aka arrive at the best accuracy).
d. An iterative approach is defined as the error for a given linear system with a precise solution. If a matrix exists, an iterative process is said to be linear. This matrix is known as the iteration matrix.
e. Examples of big mathematical models having a big potential influence on us all include general weather forecasts, global warming, flight simulation, hurricane forecasting, nuclear winter, nuclear arms race, and so on.