epsilon-exhausted version space
Suppose we are in the modelling framework with some training data
, a hypothesis space and a probability distribution on . For some the version space is -exhausted if and only if for all the true error
epsilon-exhausted version space
Suppose we are in the modelling framework with some training data
, a hypothesis space and a probability distribution on . For some the version space is -exhausted if and only if for all the true error