Avaliação da Alteração de Precisão e Leitura de Dados para Melhoria no Desempenho e Consumo de Energia de Algoritmos de Aprendizado de Máquina
Resumo
Neste trabalho é avaliada a influência da redução da precisão do ponto flutuante e de técnicas de leitura de dados nos algoritmos de Aprendizado de Máquina durante a fase de treinamento. O estudo é feito com o uso do modelo de Floresta Randômica. O objetivo é analisar aspectos de precisão, do tempo de execução e do consumo de energia, afim de realizar um melhor uso dos recursos computacionais em busca de soluções para um aprendizado de máquina ecologicamente viável.Referências
Bifet, A., Holmes, G., Kirkby, R., and Pfahringer, B. (2010). Moa: Massive online analysis http://sourceforge.net/projects/moa-datastream. Journal of Machine Learning Research (JMLR).
Breiman, L. (2001). Random forests. Machine learning, 45(1):5–32.
Breiman, L., Friedman, J., Stone, C. J., and Olshen, R. A. (1984). Classification and regression trees. CRC press.
Gron, A. (2017). Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems. O’Reilly Media, Inc., 1st edition.
Haidar, A., Wu, P., Tomov, S., and Dongarra, J. (2017). Investigating half precision arithmetic to accelerate dense linear system solvers. In Proceedings of the 8th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, pages 1–8.
Harris, C. R., Millman, K. J., van der Walt, S. J., Gommers, R., Virtanen, P., Cournapeau, D., Wieser, E., Taylor, J., Berg, S., Smith, N. J., Kern, R., Picus, M., Hoyer, S., van Kerkwijk, M. H., Brett, M., Haldane, A., del Río, J. F., Wiebe, M., Peterson, P., Gérard-Marchant, P., Sheppard, K., Reddy, T., Weckesser, W., Abbasi, H., Gohlke, C., and Oliphant, T. E. (2020). Array programming with NumPy. Nature, 585(7825):357–362.
Hashemi, S., Anthony, N., Tann, H., Bahar, R. I., and Reda, S. (2017). Understanding the impact of precision quantization on the accuracy and energy In Design, Automation Test in Europe Conference Exhibition of neural networks. (DATE), 2017, pages 1474–1479.
Holt, J. and Sievert, S. (2021). Training machine learning models faster with dask. SciPy Conferences.
Ibañez, M. M., Ramos, F. M., and , A. R. C. (2016). Uso de redes neurais nebulosas e florestas aleatórias na classificação de imagens em um projeto de ciência cidadã. Master’s thesis.
Jain, S., Venkataramani, S., Srinivasan, V., Choi, J., Chuang, P., and Chang, L. (2018). Compensated-dnn: Energy efficient low-precision deep neural networks by compensating quantization errors. In 2018 55th ACM/ESDA/IEEE Design Automation Conference (DAC), pages 1–6.
Kaggle (2020). State of data science and machine learning 2020. Technical report.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E. (2011). Scikitlearn: Machine learning in Python. Journal of ML Research, 12:2825–2830.
Schwartz, R., Dodge, J., Smith, N. A., and Etzioni, O. (2020). Green ai. Commun. ACM, 63(12):54–63.
Silva, G., Schulze, B., and Ferro, M. (2021). Performance and energy efficiency analysis of machine learning algorithms towards green ai: a case study of decision tree algorithms. Master’s thesis, National Lab. for Scientific Computing.
Thompson, N. C., Greenewald, K., Lee, K., and Manso, G. F. (2020). The computational limits of deep learning.
Villani, C., Bonnet, Y., schoenauer, m., berthet, c., levin, f., cornut, a. c., and Rondepierre, B. (2018). For a meaningful artificial intelligence: towards a french and european strategy. Conseil national du numérique.
Zervakis, G., Saadat, H., Amrouch, H., Gerstlauer, A., Parameswaran, S., and Henkel, J. (2021). Approximate computing for ml: State-of-the-art, challenges and visions. In Proceedings of the 26th Asia and South Pacific Design Automation Conference, ASPDAC ’21, page 189–196, New York, NY, USA. ACM.
Zuras, D., Cowlishaw, M., Aiken, A., Applegate, M., Bailey, D., Bass, S., Bhandarkar, D., Bhat, M., Bindel, D., Boldo, S., et al. (2008). Ieee standard for floating-point arithmetic. IEEE Std, 754(2008):1–70.
Breiman, L. (2001). Random forests. Machine learning, 45(1):5–32.
Breiman, L., Friedman, J., Stone, C. J., and Olshen, R. A. (1984). Classification and regression trees. CRC press.
Gron, A. (2017). Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems. O’Reilly Media, Inc., 1st edition.
Haidar, A., Wu, P., Tomov, S., and Dongarra, J. (2017). Investigating half precision arithmetic to accelerate dense linear system solvers. In Proceedings of the 8th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, pages 1–8.
Harris, C. R., Millman, K. J., van der Walt, S. J., Gommers, R., Virtanen, P., Cournapeau, D., Wieser, E., Taylor, J., Berg, S., Smith, N. J., Kern, R., Picus, M., Hoyer, S., van Kerkwijk, M. H., Brett, M., Haldane, A., del Río, J. F., Wiebe, M., Peterson, P., Gérard-Marchant, P., Sheppard, K., Reddy, T., Weckesser, W., Abbasi, H., Gohlke, C., and Oliphant, T. E. (2020). Array programming with NumPy. Nature, 585(7825):357–362.
Hashemi, S., Anthony, N., Tann, H., Bahar, R. I., and Reda, S. (2017). Understanding the impact of precision quantization on the accuracy and energy In Design, Automation Test in Europe Conference Exhibition of neural networks. (DATE), 2017, pages 1474–1479.
Holt, J. and Sievert, S. (2021). Training machine learning models faster with dask. SciPy Conferences.
Ibañez, M. M., Ramos, F. M., and , A. R. C. (2016). Uso de redes neurais nebulosas e florestas aleatórias na classificação de imagens em um projeto de ciência cidadã. Master’s thesis.
Jain, S., Venkataramani, S., Srinivasan, V., Choi, J., Chuang, P., and Chang, L. (2018). Compensated-dnn: Energy efficient low-precision deep neural networks by compensating quantization errors. In 2018 55th ACM/ESDA/IEEE Design Automation Conference (DAC), pages 1–6.
Kaggle (2020). State of data science and machine learning 2020. Technical report.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E. (2011). Scikitlearn: Machine learning in Python. Journal of ML Research, 12:2825–2830.
Schwartz, R., Dodge, J., Smith, N. A., and Etzioni, O. (2020). Green ai. Commun. ACM, 63(12):54–63.
Silva, G., Schulze, B., and Ferro, M. (2021). Performance and energy efficiency analysis of machine learning algorithms towards green ai: a case study of decision tree algorithms. Master’s thesis, National Lab. for Scientific Computing.
Thompson, N. C., Greenewald, K., Lee, K., and Manso, G. F. (2020). The computational limits of deep learning.
Villani, C., Bonnet, Y., schoenauer, m., berthet, c., levin, f., cornut, a. c., and Rondepierre, B. (2018). For a meaningful artificial intelligence: towards a french and european strategy. Conseil national du numérique.
Zervakis, G., Saadat, H., Amrouch, H., Gerstlauer, A., Parameswaran, S., and Henkel, J. (2021). Approximate computing for ml: State-of-the-art, challenges and visions. In Proceedings of the 26th Asia and South Pacific Design Automation Conference, ASPDAC ’21, page 189–196, New York, NY, USA. ACM.
Zuras, D., Cowlishaw, M., Aiken, A., Applegate, M., Bailey, D., Bass, S., Bhandarkar, D., Bhat, M., Bindel, D., Boldo, S., et al. (2008). Ieee standard for floating-point arithmetic. IEEE Std, 754(2008):1–70.
Publicado
26/10/2021
Como Citar
VIEIRA, Vitor; BERNARDO, Felipe; SCHULZE, Bruno; FERRO, Mariza.
Avaliação da Alteração de Precisão e Leitura de Dados para Melhoria no Desempenho e Consumo de Energia de Algoritmos de Aprendizado de Máquina. In: WORKSHOP DE INICIAÇÃO CIENTÍFICA - SIMPÓSIO EM SISTEMAS COMPUTACIONAIS DE ALTO DESEMPENHO (SSCAD), 22. , 2021, Belo Horizonte/MG.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2021
.
p. 33-40.
DOI: https://doi.org/10.5753/wscad_estendido.2021.18638.
