Table 4 Brief comparative overview on the learn to rank approaches, adapted from [ 34 ]

Ranking by learning on object pairs

Ranking by learning on object lists

pairwise loss, e.g., hinge loss, exponential loss, logistic loss

listwise loss, e.g., cross entropy loss, cosine loss

Theoretical aspects are well studied

Considers the relationship among objects to their full extent

Considers only pairwise orders; May be biased towards lists with more objects

Theoretical aspects are less well studied

SVMRank [19]; RankNet [21]; RankBoost [20]

ListNet [22]