Multilabel predictions with sets of probabilities: The Hamming and ranking loss cases

作者：

Highlights：

• We consider the problem of multilabel prediction with probability sets.

• We show that efficient approximations can be obtained for the Hamming loss.

• We show that efficient approximations can be obtained for the ranking loss.

• We perform first experiments confirming the expected behaviour of the approximations.

摘要

Highlights•We consider the problem of multilabel prediction with probability sets.•We show that efficient approximations can be obtained for the Hamming loss.•We show that efficient approximations can be obtained for the ranking loss.•We perform first experiments confirming the expected behaviour of the approximations.

论文关键词：Multilabel,Cautious predictions,Weak labels,Probability sets

论文评审过程：Received 8 November 2014, Revised 13 March 2015, Accepted 18 April 2015, Available online 29 April 2015, Version of Record 16 July 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.04.020