For linear regression Y = 00 +0X + e, i = 1,..., N minimizing squared loss...

2. 

Transcribed Image Text:

For linear regression Y = 00 +0X + e, i = 1,..., N minimizing squared loss (e; is random noise variable): a) Calculate the gradient with respect to the parameter vector. b) Write down the steps of the (batch) gradient descent rule. c) Write down the steps of the stochastic gradient descent rule. For linear regression Y = 00 +0X + e, i = 1,..., N minimizing squared loss (e; is random noise variable): a) Calculate the gradient with respect to the parameter vector. b) Write down the steps of the (batch) gradient descent rule. c) Write down the steps of the stochastic gradient descent rule.