At the energies of the Large Hardon Collider vector bosons, higgs bosons and top quarks are often produced with momenta significantly higher than their rest mass. This means that if they decay hadronically their decay products boosted such that they fall in a small area in eta-phi. The identification of these interesting objects over the large backgrounds of jets from QCD processes poses an interesting reconstruction challenge. An efficient way to reconstruct these is to use large-radius jets. As well as the mass of the large-radius jet, sub-structure variables that attempt to separate the hard multi-prong structure of these interesting jets from the QCD radiation pattern are used. Recently, machine learning techniques have been employed to fully exploit the correlations in these variables and the detector's capabilities. Measuring the efficiency of identifying these objects in data and the background rejection that can be achieved has been a primary focus such that these complicated taggers can be used in analyses. Finally the sensitivity of these tagging techniques to pile-up -- additional simultaneous collisions with the collisions of interest -- will be shown along with the various methods used to mitigate such effects.