Question: One might argue that every learning algorithm works with a fixed set of decision rules, namely, the set of all rules that the particular algorithm
One might argue that every learning algorithm works with a fixed set of decision rules, namely, the set of all rules that the particular algorithm might possibly produce over all possible observations. In light of such an argument, what are the advantages of explicitly specifying a class of rules C and studying the problem of learning using rules from this class?
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
