Learning logic programs by discovering higher-order abstractions
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916109172604928 |
|---|---|
| author | Hocquette, Céline Dumančić, Sebastijan Cropper, Andrew |
| author_facet | Hocquette, Céline Dumančić, Sebastijan Cropper, Andrew |
| contents | We introduce the higher-order refactoring problem, where the goal is to compress a logic program by discovering higher-order abstractions, such as map, filter, and fold. We implement our approach in Stevie, which formulates the refactoring problem as a constraint optimisation problem. Our experiments on multiple domains, including program synthesis and visual reasoning, show that refactoring can improve the learning performance of an inductive logic programming system, specifically improving predictive accuracies by 27% and reducing learning times by 47%. We also show that Stevie can discover abstractions that transfer to multiple domains. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2308_08334 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Learning logic programs by discovering higher-order abstractions Hocquette, Céline Dumančić, Sebastijan Cropper, Andrew Machine Learning Artificial Intelligence Programming Languages We introduce the higher-order refactoring problem, where the goal is to compress a logic program by discovering higher-order abstractions, such as map, filter, and fold. We implement our approach in Stevie, which formulates the refactoring problem as a constraint optimisation problem. Our experiments on multiple domains, including program synthesis and visual reasoning, show that refactoring can improve the learning performance of an inductive logic programming system, specifically improving predictive accuracies by 27% and reducing learning times by 47%. We also show that Stevie can discover abstractions that transfer to multiple domains. |
| title | Learning logic programs by discovering higher-order abstractions |
| topic | Machine Learning Artificial Intelligence Programming Languages |
| url | https://arxiv.org/abs/2308.08334 |