Reliable Multilabel Classification: Prediction with Partial Abstention
| Type of publication: | Article |
| Citation: | ln:MlcAbstention |
| Journal: | Proceedings of the AAAI Conference on Artificial Intelligence |
| Volume: | 34 |
| Number: | 04 |
| Year: | 2020 |
| Month: | Apr. |
| Pages: | 5264-5271 |
| URL: | https://ojs.aaai.org/index.php/AAAI/article/view/5972 |
| DOI: | 10.1609/aaai.v34i04.5972 |
| Abstract: | In contrast to conventional (single-label) classification, the setting of multilabel classification (MLC) allows an instance to belong to several classes simultaneously. Thus, instead of selecting a single class label, predictions take the form of a subset of all labels. In this paper, we study an extension of the setting of MLC, in which the learner is allowed to partially abstain from a prediction, that is, to deliver predictions on some but not necessarily all class labels. We propose a formalization of MLC with abstention in terms of a generalized loss minimization problem and present first results for the case of the Hamming loss, rank loss, and F-measure, both theoretical and experimental. |
| Keywords: | |
| Authors | |
|
Topics
|
|
|
|
