Utilize este identificador para referenciar este registo:
https://hdl.handle.net/1822/87094
Título: | AutoOC: Automated multi-objective design of deep autoencoders and one-class classifiers using grammatical evolution |
Autor(es): | Ferreira, Luís Fernando Faria Cortez, Paulo |
Palavras-chave: | Automated machine learning Deep autoencoders Grammatical evolution Multi-objective optimization One-class classification |
Data: | 2023 |
Editora: | Elsevier B.V. |
Revista: | Applied Soft Computing |
Citação: | Ferreira, L., & Cortez, P. (2023, September). AutoOC: Automated multi-objective design of deep autoencoders and one-class classifiers using grammatical evolution. Applied Soft Computing. Elsevier BV. http://doi.org/10.1016/j.asoc.2023.110496 |
Resumo(s): | One-Class Classification (OCC) corresponds to a subclass of unsupervised Machine Learning (ML) that is valuable when labeled data is non-existent. In this paper, we present AutoOC, a computationally efficient Grammatical Evolution (GE) approach that automatically searches for OCC models. AutoOC assumes a multi-objective optimization, aiming to increase the OCC predictive performance while reducing the ML training time. AutoOC also includes two execution speedup mechanisms, a periodic training sampling, and a multi-core fitness evaluation. In particular, we study two AutoOC variants: a pure Neuroevolution (NE) setup that optimizes two types of deep learning models, namely dense Autoencoder (AE) and Variational Autoencoder (VAE); and a general Automated Machine Learning (AutoML) ALL setup that considers five distinct OCC base learners, specifically Isolation Forest (IF), Local Outlier Factor (LOF), One-Class SVM (OC-SVM), AE and VAE. Several experiments were conducted, using eight public OpenML datasets and two validation scenarios (unsupervised and supervised). The results show that AutoOC requires a reasonable amount of execution time and tends to obtain lightweight OCC models. Moreover, AutoOC provides quality predictive results, outperforming a baseline IF for all analyzed datasets and surpassing the best supervised OpenML human modeling for two datasets. |
Tipo: | Artigo |
URI: | https://hdl.handle.net/1822/87094 |
DOI: | 10.1016/j.asoc.2023.110496 |
ISSN: | 1568-4946 |
Versão da editora: | https://www.sciencedirect.com/science/article/pii/S1568494623005148 |
Arbitragem científica: | yes |
Acesso: | Acesso aberto |
Aparece nas coleções: | CAlg - Artigos em revistas internacionais / Papers in international journals |
Ficheiros deste registo:
Ficheiro | Descrição | Tamanho | Formato | |
---|---|---|---|---|
AutoOC.pdf | 508,48 kB | Adobe PDF | Ver/Abrir |