超越具体化：重思西蒙东的技术有机体演化——基于卷积神经网络的考察
Transcending Concretization: Rethinking Simondon’s Theory of Technological Evolution—An Investigation Based on Convolutional Neural Networks
摘要: 人工智能技术的迅速发展使思考技术演进之逻辑愈发必要。西蒙东的具体化理论主张技术演化在于功能与结构的逐步协调，在这一过程中技术趋向于有机。然而，该理论未能有效地解释现代人工智能技术的发展。本文以卷积神经网络为考察对象，梳理了早期LeNet至深度学习时代AlexNet的演进路径。研究表明，人工智能技术的演进往往优先考虑计算效率、并行性、软硬件协同等性能提升，而非单纯的具体化，有时甚至表现出“去具体化”趋势。由此，我们将超越西蒙东的具体化路径并试图揭示出一种超越“人类中心主义”的技术演进逻辑。
Abstract: The rapid development of artificial intelligence technology makes it increasingly necessary to contemplate the logic of technological evolution. Simondon’s concretization theory posits that technological evolution lies in the gradual coordination of function and structure, during which technology tends to become organic. However, this theory fails to adequately explain the development of modern artificial intelligence technology. This paper examines convolutional neural networks by tracing the evolutionary path from the early LeNet to AlexNet in the deep learning era. The study shows that the evolution of artificial intelligence technology often prioritizes improvements in performance—such as computational efficiency, parallelism, and the synergy between software and hardware—rather than mere concretization, sometimes even exhibiting a trend toward “deconcretization.” Consequently, we aim to go beyond Simondon’s concretization pathway and attempt to unveil a logic of technological evolution that transcends anthropocentrism.
文章引用：王汉瑞. 超越具体化：重思西蒙东的技术有机体演化——基于卷积神经网络的考察[J]. 哲学进展, 2025, 14(5): 210-216. https://doi.org/10.12677/acpp.2025.145234

参考文献

[1] 吉尔贝·西蒙东. 论技术物的存在模式[M]. 南京: 南京大学出版社, 2024: 34, 50, 32, 22, 95.
[2] Lindberg, S. (2019) Being with Technique-Technique as Being-With: The Technological Communities of Gilbert Simondon. Continental Philosophy Review, 52, 299-310. [Google Scholar] [CrossRef] 
[3] Chellapilla, K., Puri, S. and Simard, P. (2006) High Performance Convolutional Neural Networks for Document Processing. 10th International Workshop on Frontiers in Handwriting Recognition, La Baule, 23-26 October 2006.
[4] Raina, R., Madhavan, A. and Ng, A.Y. (2009) Large-Scale Deep Unsupervised Learning Using Graphics Processors. Proceedings of the 26th Annual International Conference on Machine Learning, Vol. 9, 873-880. [Google Scholar] [CrossRef] 
[5] LeCun, Y., Bengio, Y. and Hinton, G. (2015) Deep Learning. Nature, 521, 436-444. [Google Scholar] [CrossRef] [PubMed]
[6] Lecun, Y., Bottou, L., Bengio, Y. and Haffner, P. (1998) Gradient-Based Learning Applied to Document Recognition. Proceedings of the IEEE, 86, 2278-2324. [Google Scholar] [CrossRef] 
[7] Ciresan, D.C., Meier, U., Masci, J., et al. (2011) Flexible, High Performance Convolutional Neural Networks for Image Classification. IJCAI Proceedings-International Joint Conference on Artificial Intelligence, 22, 1237.
[8] Krizhevsky, A., Sutskever, I. and Hinton, G.E. (2012) ImageNet Classification with Deep Convolutional Neural Networks. Communications of the ACM, 60, 84-90.
[9] Ghimire, D., Kil, D. and Kim, S. (2022) A Survey on Efficient Convolutional Neural Networks and Hardware Acceleration. Electronics, 11, Article No. 945. [Google Scholar] [CrossRef] 
[10] 斯图亚特·罗素. 人工智能现代方法[M]. 第四版. 北京: 人民邮电出版社, 2022.
[11] Nair, V. and Hinton, G.E. (2010) Rectified Linear Units Improve Restricted Boltzmann Machines. Proceedings of the 27th International Conference on Machine Learning (ICML-10), Haifa, 21-24 June 2010, 807-814.
[12] Pfeifer, R. and Iida, F. (2004) Embodied Artificial Intelligence: Trends and Challenges. In: Iida, F., et al., Eds., Embodied Artificial Intelligence, Springer, 1-26. [Google Scholar] [CrossRef] 
[13] Stiegler, B. (1998) Technics and Time, 1: The Fault of Epimetheus. Trans. Richard Beardsworth and George Collins, Stanford University Press, 152.

为你推荐