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

Abstract : In this paper, we study how multilabel predictions can be obtained when our uncertainty is described by a convex set of probabilities. Such predictions, typically consisting of a set of potentially optimal decisions, are hard to make in large decision spaces such as the one considered in multilabel problems. However, we show that when considering the Hamming or the ranking loss, outer-approximating predictions can be efficiently computed from label-wise information, as in the precise case. We also perform some first experiments showing the behaviour of the partial predictions obtained through these approximations. Such experiments also confirm that predictions become partial on those labels where the precise prediction is likely to make an error.
Document type :
Journal articles
Complete list of metadatas

Cited literature [37 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01179430
Contributor : Sébastien Destercke <>
Submitted on : Wednesday, July 22, 2015 - 2:40:08 PM
Last modification on : Tuesday, July 24, 2018 - 4:40:02 PM
Long-term archiving on : Friday, October 23, 2015 - 11:11:56 AM

File

Notes_els.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Sébastien Destercke. Multilabel predictions with sets of probabilities: The Hamming and ranking loss cases. Pattern Recognition, Elsevier, 2015, pp.3757-3765. ⟨10.1016/j.patcog.2015.04.020⟩. ⟨hal-01179430⟩

Share

Metrics

Record views

191

Files downloads

473