The section in unit 6 on regularization needs a lot of work. It is a bit out of sync with how I explain in class, which was heavily inspired by this text. In particular, it should first include the simple example of (ridge) regularization for a sample mean, then do the same for linear regression, noting how the parameter \lambda is related to the prior variance in Bayesian inference
The section in unit 6 on regularization needs a lot of work. It is a bit out of sync with how I explain in class, which was heavily inspired by this text. In particular, it should first include the simple example of (ridge) regularization for a sample mean, then do the same for linear regression, noting how the parameter \lambda is related to the prior variance in Bayesian inference