Section: New Software and Platforms

Ranking with High-Order Information

Participant : Puneet Dokania [Correspondant] .

Average precision optimization, high-order information, ranking The software (publicly available at http://cvn.ecp.fr/projects/ranking-highorder/ ) provides a convenient API for learning to rank with high-order information. The samples are ranked according to a scorethat is proportional to the difference of max-marginals of the positive and the negative class. The parameters of the score function are computed by minimizing an upper bound on the average precision loss. The software also provides an instantiation of the API for ranking samples according to their relevance to an action, using the poselet features.