Question: a Machine Learning algorithm uses a finite hypothesis classH .h M) with M 1024 to select a hypothesis g e H with the smallest training
a Machine Learning algorithm uses a finite hypothesis classH .h M) with M 1024 to select a hypothesis g e H with the smallest training error LD (g) among all hypotheses in H. Let L() be the true risk of g. Give the smallest number of data points N required to ensure that with probability at least 0.99 we have 14 (g)-LD(g)| 0.05
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
