Question

If the features of model 1 are a strict subset of those in model 2, which model will usually have lowest training error?

Answer 1

remark question as brain list

Answer 2

Answer: Model 2 has low training error

Explanation:

Because as model 2 is super set of model 1 =====> that model 2 is more complex than model 1 and as we know that more complex model fit better than simple model therefore they have high variance and less bias like Decision Trees.