Learning logic programs by discovering higher-order abstractions
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 experiment...
| Main Authors: | , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
International Joint Conferences on Artificial Intelligence
2024
|
| _version_ | 1824458730868047872 |
|---|---|
| author | Hocquette, C Dumancic, S Cropper, A |
| author_facet | Hocquette, C Dumancic, S Cropper, A |
| author_sort | Hocquette, C |
| collection | OXFORD |
| description | 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. |
| first_indexed | 2025-02-19T04:30:32Z |
| format | Conference item |
| id | oxford-uuid:c4ecf3e8-d5ef-4af7-b489-345b5cf391a7 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2025-02-19T04:30:32Z |
| publishDate | 2024 |
| publisher | International Joint Conferences on Artificial Intelligence |
| record_format | dspace |
| spelling | oxford-uuid:c4ecf3e8-d5ef-4af7-b489-345b5cf391a72024-12-17T09:07:39ZLearning logic programs by discovering higher-order abstractionsConference itemhttp://purl.org/coar/resource_type/c_5794uuid:c4ecf3e8-d5ef-4af7-b489-345b5cf391a7EnglishSymplectic ElementsInternational Joint Conferences on Artificial Intelligence2024Hocquette, CDumancic, SCropper, AWe 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. |
| spellingShingle | Hocquette, C Dumancic, S Cropper, A Learning logic programs by discovering higher-order abstractions |
| title | Learning logic programs by discovering higher-order abstractions |
| title_full | Learning logic programs by discovering higher-order abstractions |
| title_fullStr | Learning logic programs by discovering higher-order abstractions |
| title_full_unstemmed | Learning logic programs by discovering higher-order abstractions |
| title_short | Learning logic programs by discovering higher-order abstractions |
| title_sort | learning logic programs by discovering higher order abstractions |
| work_keys_str_mv | AT hocquettec learninglogicprogramsbydiscoveringhigherorderabstractions AT dumancics learninglogicprogramsbydiscoveringhigherorderabstractions AT croppera learninglogicprogramsbydiscoveringhigherorderabstractions |