Desafíos epistemológicos en la era de las redes neuronales artificiales: abordando sistemas complejos desde una perspectiva computacional

• Autores: Erick Manuel Rubio, Leandro Giri, Andrés A. Ilcic
• Localización: Argumentos de razón técnica: Revista española de ciencia, tecnología y sociedad, y filosofía de la tecnología, ISSN 1139-3327, Nº 26, 2023, págs. 145-178
• Idioma: español
• Títulos paralelos:
  • Epistemological challenges in the era of artificial neural networks: addressing complex systems from a computational perspective
• Enlaces
  • Texto completo  1    2  
• Dialnet Métricas: 1 Cita
• Resumen
  • español

    En este artículo se explora un enfoque “manipulacionista” en la investigación científica, utilizando redes neuronales artificiales para abordar sistemas complejos. Se destaca la capacidad de estas redes para resolver problemas y representar patrones de comportamiento en estos sistemas. Además, se enfatiza la importancia de superar las limitaciones de los métodos analíticos tradicionales mediante métodos computacionales y la visualización de grandes cantidades de datos. El artículo toma como ejemplo significativo el problema del plegamiento de proteínas y el éxito de AlphaFold2 en este campo. Se plantea la necesidad de explorar las implicaciones epistemológicas más amplias de las simulaciones computacionales en la investigación científica, considerando la relación entre precisión predictiva y comprensión de los sistemas complejos estudiados. La estructura del trabajo se divide en cuatro secciones, que incluyen el estudio de sistemas complejos, el problema del plegamiento de proteínas y una discusión filosófica sobre los modelos de simulación computacional y las redes neuronales artificiales; además de una sección de comentarios finales. En conjunto, se enfatiza la relevancia de este enfoque para abordar sistemas complejos y su impacto en la filosofía de la ciencia.

  • English

    The article delves into a “manipulationist”approach in scientific research, utilizing artificial neural networks to tackle complex systems. It highlights the networks' capacity to resolve problems and portray behavioral patterns within these intricate systems. Moreover, it underscores the significance of surmounting the limitations posed by traditional analytical methods through computational techniques and the visualization of vast amounts of data. The article specifically focuses on the issue of protein folding and mentions the achievements of systems like AlphaFold2 in this domain. It highlights the necessity to explore the broader epistemological implications of computational simulations in scientific research, taking into account the interplay between predictive accuracy and the comprehension of the studied complex systems. The work is structured into four sections, encompassing the study of complex systems, the problem of protein folding, a philosophical discussion on computational simulation models and artificial neural networks, and concluding with final remarks. Overall, it underscores the significance of this approach in addressing complex systems and its profound impact on the philosophy of science.

• Referencias bibliográficas
  • Barak, O. (2017). Recurrent neural networks as versatile tools of neuroscience research. Current Opinion in Neurobiology, 46, 1-6.
  • Bunge, M. (2004). Emergencia y Convergencia. Novedad Cualitativa y Unidad del Conocimiento (Rafael González del Solar, Trad.). Gedisa. (Obra...
  • Chirimuuta, M. (2020). Prediction versus understanding in computational enhanced neuroscience. Synthese, 199(4).
  • Churchland, P. S., Sejnowski, T. J. (2016). Blending computational and experimental neuroscience. Nature Reviews Neuroscience, 17, 667–668.
  • Durán, J. M. (2018). Computer Simulations in Science and Engineering: Concepts -Practice -Perspectives. Springer Nature Switzerland.
  • Durán, J. M., Jongsma, K. R. (2021). Who is afraid of black box algorithms? On the epistemological and ethical basis of trust in medical...
  • Evans, R., Jumper, J., Kirkpatrick, J., Sifre, L., Green, T., Qin, C., …, Senior, A. (2018). De novo structure prediction with deeplearning...
  • Feurer, M., Hutter, F. (2019). Hyperparameter Optimization, en: Hutter F., Kotthoff, L., & Vanschoren, J. (eds.) Automated Machine Learning:...
  • Frigg, R. (2022). Models and Theories: a Philosophical Inquiry. Routledge.
  • Galison, P. (1996). Computer Simulations and the Trading Zone, en: Galison P., & Stump, J. D. (eds). The Disunity of Science: Boundaries,...
  • Gao, P., Ganguli, S. (2015). On simplicity and complexity in the brave new world of large-scale neuroscience. Current Opinion in Neurobiology,...
  • Gell-Mann, M. (1998). El Quark y el Jaguar: aventuras en lo simple y lo complejo (A. García y R. Pastor, Trad.). Tusquets. (Obra original...
  • Goldenfeld, N. & Kadanoff, L. P. (1999). Simple lessons from complexity. Science, 284: 87- 89.
  • Hartmann, S. (1996). The World as a Process: Simulations in the Natural and Social Sciences, En: Hegselmann, R., Mueller, U., Troitzsch,...
  • Hawkins, J., Blakeslee, S. (2005). On Intelligence: How a New Understanding of the Brain Will Lead to the Creation of Truly Intelligent Machines....
  • Hidalgo, C. (2015). Why Information Grows. The Evolution of Order, from Atoms to Economies. New York: Basic Book.
  • Holland, J. H. (1992). Complex Adaptive Systems. Daedalus, 121(1), 17–30.
  • Humphreys, P. (1997). Emergence, Not Supervenience. Philosophy of Science, 64, 337- 345.
  • Humphreys, P. (2004). Extending Ourselves. Computational Sciences, Empiricism, and Scientific Method. Oxford University
  • Humphreys, P. (2009). The philosophical novelty of computer simulation methods. Synthese, 169, 615-626.
  • Humphreys, P. (2016). Emergence: A Philosophical Account. Oxford University Press.
  • Johnson, S. (2003). Sistemas emergentes. O qué tienen en común hormigas, neuronas, ciudades y software. Fondo de Cultura Económica.
  • Jumper, J., Evans, R., Pritzel, A. Green, T., Figurnov, M., Ronneberger, O., …, Hassabis, D. (2021). Highly accurate protein structure prediction...
  • Kauffman, S. A. (1993). The Origins of Order: Self-organization and Selection in Evolution. Oxford University Press.Laughlin, R. B.,
  • Pines, D. (2000). The Theory of Everything. Proceedings of the National Academy of Sciences, 97(1), 28-31.
  • Lenhard, J. (2016). Computer Simulation, en: Humphreys, P. (ed.) The Oxford Handbook of Philosophy of Science (pp. 803-824). Oxford University...
  • Lenhard, J. (2017). License to Explore: How Images Work in Simulation Modeling, en: Ammon S., & Capdevila-Werning, R. (eds). The Active...
  • Levinthal, C. (1968). Are there pathways for protein folding? Journal de Chimie Physique, 65, 44-45.
  • Lillicrap, T., Kording, K. (2019). What does it mean to understand a neural network?
  • Lillicrap, T. P., Santoro, A., Marris, L., Akerman, C. J., Hinton, G. (2020). Backpropagation and the brain. Nature Reviews Neuroscience,...
  • Mitchell, M. (2009). Complexity. A Guided Tour. Oxford University
  • Napoletani, D., Panza, M., Struppa, D. C. (2011). Agnostic science. Towards a philosophy of data analysis. Foundations of Science, 16(1),...
  • Nicolis, G., Prigogine, I. (1989). Exploring Complexity. An Introduction. W. H. Freeman & Company.
  • Nielsen, M. (2015). Neural Networks and Deep Learning.
  • Pearl, J., Mackenzie, D. (2018). The Book of Why: The New Science of Cause and Effect. Basic Books.
  • Puterman, M. (1990). Markov Decision Processes. En Handbooks in Operations Research and Management Science Vol 2 (pp. 331-434). Elsevier.
  • Rescher, N. (1998). Complexity. A Philosophical Overview. Transaction Publishers.
  • Rumelhart, D., Hinton, G., Williams, R. (1986). Learning representations by back-propagation errors. Nature, 323, 533 536
  • Senior A.W., Evans, R., Jumper, J., Kirkpatrick, J., Sifre, L., Green, T., …, Hassabis, D. (2020). Improved protein structure prediction using...
  • Taylor, M. C. (2001). The Moment of Complexity. Emerging Network Culture. Chicago: The University of Chicago Press.
  • Vicsek, T. (2002). The Bigger Picture. Nature, 418: 131.
  • Weiskopf, D. (2022). The Predictive Turn in Neuroscience. Philosophy of Science (5).
  • Woodward, J. (2003). Making Things Happen. A theory of causal explanation. New York: Oxford University Press.
  • Yang, L., Shami, A. (2020). On hyperparameter optimization of machine learning algorithms: Theory and practice. Neurocomputing, 415: 295-316
  • Zednik, C. (2019). Solving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence. arXiv:1903.04361 [cs].