A fuzzy loss for ontology classification
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910570382360576 |
|---|---|
| author | Flügel, Simon Glauer, Martin Mossakowski, Till Neuhaus, Fabian |
| author_facet | Flügel, Simon Glauer, Martin Mossakowski, Till Neuhaus, Fabian |
| contents | Deep learning models are often unaware of the inherent constraints of the task they are applied to. However, many downstream tasks require logical consistency. For ontology classification tasks, such constraints include subsumption and disjointness relations between classes.
In order to increase the consistency of deep learning models, we propose a fuzzy loss that combines label-based loss with terms penalising subsumption- or disjointness-violations. Our evaluation on the ChEBI ontology shows that the fuzzy loss is able to decrease the number of consistency violations by several orders of magnitude without decreasing the classification performance. In addition, we use the fuzzy loss for unsupervised learning. We show that this can further improve consistency on data from a |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_02083 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A fuzzy loss for ontology classification Flügel, Simon Glauer, Martin Mossakowski, Till Neuhaus, Fabian Artificial Intelligence Logic in Computer Science Deep learning models are often unaware of the inherent constraints of the task they are applied to. However, many downstream tasks require logical consistency. For ontology classification tasks, such constraints include subsumption and disjointness relations between classes. In order to increase the consistency of deep learning models, we propose a fuzzy loss that combines label-based loss with terms penalising subsumption- or disjointness-violations. Our evaluation on the ChEBI ontology shows that the fuzzy loss is able to decrease the number of consistency violations by several orders of magnitude without decreasing the classification performance. In addition, we use the fuzzy loss for unsupervised learning. We show that this can further improve consistency on data from a |
| title | A fuzzy loss for ontology classification |
| topic | Artificial Intelligence Logic in Computer Science |
| url | https://arxiv.org/abs/2405.02083 |