La Aritmética del Futurouna Reflexión sobre los Planes de Estudio
- Raul Murillo
- Alberto A. Del Barrio
- Guillermo Botella
ISSN: 2173-8688
Year of publication: 2020
Issue: 10
Pages: 49-60
Type: Article
More publications in: Enseñanza y aprendizaje de ingeniería de computadores: Revista de Experiencias Docentes en Ingeniería de Computadores
Abstract
Con los nuevos planes de estudio, el número de horas dedicadas a la Aritmética de Computadores ha disminuido de forma notable. Sin embargo, con la explosión en la última década de las técnicas de Machine Learning y sobre todo las redes neuronales, nuevos formatos aritméticos han entrado en escena. En este trabajo hacemos un repaso de los más significativos, especialmente de los posits, introducidos en 2017 por John L. Gustafson como un reemplazo directo del punto flotante. Por tanto consideramos que en los futuros planes de estudio, la Aritmética de Computadores tiene que estar presente para garantizar que los nuevos ingenieros puedan trabajar exitosamente con estos nuevos formatos en aplicaciones críticas como las redes neuronales.
Bibliographic References
- 1. Ercegovac, M.D., Lang, T.: Digital arithmetic. Elsevier, 1st edn. (2004)
- 2. Koren, I.: Computer Arithmetic Algorithms. A. K. Peters, Ltd., USA, 2nd edn. (2002)
- 3. Hennessy, J.L., Patterson, D.A.: Computer architecture: a quantitative approach. Morgarn Kaufmann, 5th edn. (2011)
- 4. Hennessy, J.L., Patterson, D.A.: A new golden age for computer architecture. Communications of the ACM 62(2), 48–60 (2019)
- 5. LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proceedings of the IEEE 86(11), 2278–2324 (1998). https://doi.org/10.1109/5.726791
- 6. Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems 25. pp. 1097–1105 (2012)
- 7. Johnson, J.: Rethinking floating point for deep learning. ArXiv abs/1811.01721 (2018)
- 8. IEEE standard for binary floating-point arithmetic. ANSI/IEEE Std 754-1985 pp. 1–20 (Oct 1985). https://doi.org/10.1109/IEEESTD.1985.82928
- 9. Del Barrio, A.A., Bagherzadeh, N., Hermida, R.: Ultra-low-power adder stage design for exascale floating point units. ACM Transactions on Embedded Computing Systems (TECS) 13(3s), 1–24 (2014)
- 10. Kalamkar, D.D., Mudigere, D., Mellempudi, N., Das, D., Banerjee, K., Avancha, S., Vooturi, D.T., Jammalamadaka, N., Huang, J., Yuen, H., Yang, J., Park, J., Heinecke, A., Georganas, E., Srinivasan, S., Kundu, A., Smelyanskiy, M., Kaul, B., Dubey, P.: A study of bfloat16 for deep learning training. ArXiv abs/1905.12322 (2019)
- 11. Kim, M.S., Del Barrio, A.A., Oliveira, L.T., Hermida, R., Bagherzadeh, N.: Efficient mitchell’s approximate log multipliers for convolutional neural networks. IEEE Transactions on Computers 68(5), 660–675 (2018)
- 12. Kim, H., Kim, M.S., Del Barrio, A.A., Bagherzadeh, N.: A cost-efficient iterative truncated logarithmic multiplication for convolutional neural networks. In: 2019 IEEE 26th Symposium on Computer Arithmetic (ARITH). pp. 108–111. IEEE (2019)
- 13. Gustafson, J.L., Yonemoto, I.T.: Beating floating point at its own game: Posit arithmetic. Supercomputing Frontiers and Innovations 4(2), 71–86 (2017)
- 14. Posit arithmetic. https://posithub.org/docs/Posits4.pdf, [Online; ´ultimo acceso 01-06-2020]
- 15. Planes de estudio de los grados en inform´atica (UCM). https://informatica.ucm.es/grado, [Online; ´ultimo acceso 03-06-2020]
- 16. Planes de estudio de los grados en inform´atica (UPM). https://www.fi.upm.es/, [Online; ´ultimo acceso 03-06-2020]
- 17. Planes de estudio de los grados en inform´atica (UPC). https://www.upc.edu/es/grados/ingenieria-informatica-barcelona-fib, [Online; ´ultimo acceso 03-06-2020]
- 18. Planes de estudio de los grados en inform´atica (UGR). https://grados.ugr.es/informatica/, [Online; ´ultimo acceso 03-06-2020]
- 19. Reinalda, B., Kulesza, E.: The Bologna Process–Harmonizing Europe’s Higher Education: Including the Essential Original Texts. Verlag Barbara Budrich (2006)
- 20. Del Barrio, A.A., Manzano, J.P., Maroto, V.M., Villar´ın, A., Pag´an, J., Zapater, M., Ayala, J., Hermida, R.: HackRF+ GNU Radio: A software-defined radio to teach communication theory. The International Journal of Electrical Engineering & Education pp. 1–18 (2019)
- 21. Del Barrio, A.A., Hermida, R., Ogrenci-Memik, S.: A combined arithmetic-highlevel synthesis solution to deploy partial carry-save radix-8 booth multipliers in datapaths. IEEE Transactions on Circuits and Systems I: Regular Papers 66(2), 742–755 (2019)
- 22. Del Barrio, A.A., Hermida, R., Memik, S.O.: A partial carry-save on-the-fly correction multispeculative multiplier. IEEE Transactions on Computers 65(11), 3251– 3264 (2016)
- 23. Carmichael, Z., Langroudi, H.F., Khazanov, C., Lillie, J., Gustafson, J.L., Kudithipudi, D.: Deep positron: A deep neural network using the posit number system. In: 2019 Design, Automation & Test in Europe Conference & Exhibition (DATE). pp. 1421–1426. IEEE (2019)
- 24. Murillo, R., Barrio, A.A.D., Juan, G.B.: Template-based posit multiplication for training and inferring in neural networks. ArXiv abs/1907.04091 (2019)
- 25. Langroudi, H.F., Carmichael, Z., Kudithipudi, D.: Deep learning training on the edge with low-precision posits. ArXiv abs/1907.13216 (2019)
- 26. Murillo, R., Del Barrio, A.A., Botella, G.: Customized posit adders and multipliers using the flopoco core generator. In: 2020 IEEE International Symposium on Circuits and Systems (ISCAS) (2020)
- 27. Murillo, R., Del Barrio, A.A., Botella, G.: Deep pensieve: A deep learning framework based on the posit number system. Digital Signal Processing p. 102762 (2020)
- 28. New approach could sink floating-point computation. https://www.nextplatform.com/2019/07/08/new-approach-could-sink-floating-point-computation/, [Online;ultimo acceso 03-06-2020]
- 29. Tiwari, S., Gala, N., Rebeiro, C., Kamakoti, V.: Peri: A posit enabled RISC-Vcore. ArXiv abs/1908.01466 (2019)
- 30. Arunkumar, M., Bhairathi, S.G., Hayatnagarkar, H.G.: Perc: Posit enhanced rocket chip. In: Fourth Workshop on Computer Architecture Research with RISC-V at International Symposium on Computer Architecture (ISCA) (2020)
- 31. IEEE standard for floating-point arithmetic. IEEE Std 754-2008 pp. 1–70 (Aug 2008). https://doi.org/10.1109/IEEESTD.2008.4610935
- 32. Cococcioni, M., Rossi, F., Ruffaldi, E., Saponara, S.: Fast approximations of activation functions in deep neural networks when using posit arithmetic. Sensors 20(5), 1515 (2020)
- 33. Uguen, Y., Forget, L., de Dinechin, F.: Evaluating the hardware cost of the posit number system. In: 2019 29th International Conference on Field Programmable Logic and Applications (FPL). pp. 106–113. IEEE (2019)
- 34. De Dinechin, F., Forget, L., Muller, J.M., Uguen, Y.: Posits: the good, the bad and the ugly. In: Proceedings of the Conference for Next Generation Arithmetic 2019. pp. 1–10 (2019)
- 35. Krizhevsky, A.: Learning multiple layers of features from tiny images. Master’s thesis, University of Toronto, Canada (2009)