Speeding up parameter and rule learning for acyclic probabilistic logic programs
| Autor: | Francisco Henrique Otte Vieira de Faria, Fabio Gagliardi Cozman, Arthur Colombini Gusmão, Glauber De Bona, Denis Deratani Mauá |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Computer Science::Machine Learning
Theoretical computer science Computer science Applied Mathematics Orders of magnitude (acceleration) Maximum likelihood Probabilistic logic 02 engineering and technology Theoretical Computer Science Artificial Intelligence 020204 information systems 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Focus (optics) APRENDIZADO COMPUTACIONAL Structure learning Software |
| Zdroj: | Repositório Institucional da USP (Biblioteca Digital da Produção Intelectual) Universidade de São Paulo (USP) instacron:USP |
| ISSN: | 0888-613X |
| DOI: | 10.1016/j.ijar.2018.12.012 |
| Popis: | This paper introduces techniques that speed-up parameter and rule learning for acyclic probabilistic logic programs. We focus on maximum likelihood estimation of parameters, and show that significant improvements can be obtained by efficiently handling probabilistic rules. We then move to structure learning, where we learn sets of rules, by introducing an algorithm that greatly simplifies exact score-based learning. Experiments demonstrate that our methods can produce orders of magnitude speed-ups over the state-of-art in parameter and rule learning. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect