SOLVED: Gradient descent for linear regression; we calculated the following update step for every θ: θ = θ α(1/m)∑(h(x) - y)x. As you know, this rule is called the least mean squares update rule. In your own words, why should we set α to a value that is not too large?(a) Explain briefly your answer.(b) Also explain briefly why should we set α to a value that is not too small.(c) Explain why we are not worried for linear regression that this update rule might stop in a local minimum val

Gradient Descent in Machine Learning: Python Examples