Interpretabilidade de modelos de aprendizado de máquina por meio de uma rede de autoencoders
Resumo
Nos últimos anos com o aumento na capacidade dos computadores para processar cálculos vetoriais, com o aparecimento de novos algoritmos para aprendizagem de máquina, modelos de aprendizado profundo tornaram-se mais comuns. Estes modelos que são usados nas mais diferentes áreas de negócios podem ser considerados como uma caixa preta, impossível de serem depurados ou auditados. Neste trabalho vamos introduzir uma arquitetura baseada em árvore que, por ser hierárquica, possibilita ser mais transparente e interpretável. O trabalho apresentado está em fase inicial e os resultados obtidos até então são ainda preliminares.
Referências
Ian H Witten. Data mining. 2016.
Manuel Fernández-Delgado, Eva Cernadas, Senén Barro, and Dinani Amorim. Do we need hundreds of classifiers to solve real world classification problems? The journal of machine learning research, 15(1):3133–3181, 2014.
Vishnu Subramanian. Deep Learning with PyTorch: A practical approach to building neural network models using PyTorch. Packt Publishing Ltd, 2018.
Basavarajauah, M.6 basic things to know about Convolution. 2019. Disponível em: <https://medium.com/@bdhuma/6-basic-things-to-know-about-convolution-daef5e1bc411>.
Birla, D.Basics of Autoencoders. 2019. Disponível em: <https://medium.com/@birla.deepak26/autoencoders-76bb49ae6a8f>.
XIAO, Han; RASUL, Kashif; VOLLGRAF, Roland. Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747, 2017.
Riccardo Miotto, Fei Wang, Shuang Wang, Xiaoqian Jiang, Joel T Dudley, Deep learning for healthcare: review, opportunities and challenges, Briefings in Bioinformatics, Volume 19, Issue 6, November 2018, Pages 1236–1246, https://doi.org/10.1093/bib/bbx044
Muhammad Aurangzeb Ahmad, Carly Eckert, and Ankur Teredesai. 2018. Interpretable Machine Learning in Healthcare. In Proceedings of the 2018 ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (BCB '18). Association for Computing Machinery, New York, NY, USA, 559–560. DOI: https://doi.org/10.1145/3233547.3233667
S. Albawi, T. A. Mohammed and S. Al-Zawi, "Understanding of a convolutional neural network," 2017 International Conference on Engineering and Technology (ICET), 2017, pp. 1-6, doi: 10.1109/ICEngTechnol.2017.8308186.
J. Liang and R. Liu, "Stacked denoising autoencoder and dropout together to prevent overfitting in deep neural network," 2015 8th International Congress on Image and Signal Processing (CISP), 2015, pp. 697-701, doi: 10.1109/CISP.2015.7407967.
Badr, Will. Auto-Encoder: What Is It? And What Is It Used For? (Part 1), 2019. Disponível em: [link].
Manuel Fernández-Delgado, Eva Cernadas, Senén Barro, and Dinani Amorim. Do we need hundreds of classifiers to solve real world classification problems? The journal of machine learning research, 15(1):3133–3181, 2014.
Vishnu Subramanian. Deep Learning with PyTorch: A practical approach to building neural network models using PyTorch. Packt Publishing Ltd, 2018.
Basavarajauah, M.6 basic things to know about Convolution. 2019. Disponível em: <https://medium.com/@bdhuma/6-basic-things-to-know-about-convolution-daef5e1bc411>.
Birla, D.Basics of Autoencoders. 2019. Disponível em: <https://medium.com/@birla.deepak26/autoencoders-76bb49ae6a8f>.
XIAO, Han; RASUL, Kashif; VOLLGRAF, Roland. Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747, 2017.
Riccardo Miotto, Fei Wang, Shuang Wang, Xiaoqian Jiang, Joel T Dudley, Deep learning for healthcare: review, opportunities and challenges, Briefings in Bioinformatics, Volume 19, Issue 6, November 2018, Pages 1236–1246, https://doi.org/10.1093/bib/bbx044
Muhammad Aurangzeb Ahmad, Carly Eckert, and Ankur Teredesai. 2018. Interpretable Machine Learning in Healthcare. In Proceedings of the 2018 ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (BCB '18). Association for Computing Machinery, New York, NY, USA, 559–560. DOI: https://doi.org/10.1145/3233547.3233667
S. Albawi, T. A. Mohammed and S. Al-Zawi, "Understanding of a convolutional neural network," 2017 International Conference on Engineering and Technology (ICET), 2017, pp. 1-6, doi: 10.1109/ICEngTechnol.2017.8308186.
J. Liang and R. Liu, "Stacked denoising autoencoder and dropout together to prevent overfitting in deep neural network," 2015 8th International Congress on Image and Signal Processing (CISP), 2015, pp. 697-701, doi: 10.1109/CISP.2015.7407967.
Badr, Will. Auto-Encoder: What Is It? And What Is It Used For? (Part 1), 2019. Disponível em: [link].
Publicado
25/10/2021
Como Citar
CERQUEIRA, Hélisson Oliveira Magalhães; MATOS, Leonardo Nogueira.
Interpretabilidade de modelos de aprendizado de máquina por meio de uma rede de autoencoders. In: ESCOLA REGIONAL DE COMPUTAÇÃO BAHIA, ALAGOAS E SERGIPE (ERBASE), 21. , 2021, Maceió.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2021
.
p. 28-36.
DOI: https://doi.org/10.5753/erbase.2021.20053.
