Towards a Joint Approach to Produce Decisions and Explanations Using CNNs

• Tipo: Artigo em Livro de Atas de Conferência Internacional
• Data: 2020

Autores

Isabel Rio-Torto (Autor) Outra; Kelwin Fernandes (Autor) Outra; Luís F. Teixeira (Autor) FEUP

Ata de Conferência Internacional

Título: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Páginas: 3-15

9th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2019

1 July 2019 through 4 July 2019

Indexação

ISI Web of Knowledge - 0 Citações

Scopus - 1 Citação

Outras Informações

ID Authenticus: P-00R-FAP

DOI: 10.1007/978-3-030-31332-6_1

Abstract (EN): Convolutional Neural Networks, as well as other deep learning methods, have shown remarkable performance on tasks like classification and detection. However, these models largely remain black-boxes. With the widespread use of such networks in real-world scenarios and with the growing demand of the right to explanation, especially in highly-regulated areas like medicine and criminal justice, generating accurate predictions is no longer enough. Machine learning models have to be explainable, i.e., understandable to humans, which entails being able to present the reasons behind their decisions. While most of the literature focuses on post-model methods, we propose an in-model CNN architecture, composed by an explainer and a classifier. The model is trained end-to-end, with the classifier taking as input not only images from the dataset but also the explainer¿s resulting explanation, thus allowing for the classifier to focus on the relevant areas of such explanation. We also developed a synthetic dataset generation framework, that allows for automatic annotation and creation of easy-to-understand images that do not require the knowledge of an expert to be explained. Promising results were obtained, especially when using L1 regularisation, validating the potential of the proposed architecture and further encouraging research to improve the proposed architecture¿s explainability and performance. © 2019, Springer Nature Switzerland AG.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 13