[1]李德毅,刘玉超,鲍泓,等.读懂爱因斯坦——“四要素说”是统揽人和机器认知的第一性原理[J].智能系统学报,2025,20(4):1046-1052.[doi:10.11992/tis.202506028]
LI Deyi,LIU Yuchao,BAO Hong,et al.Understanding Albert Einstein: The “Four Elements Theory” is the first principle that governs human and machine cognition[J].CAAI Transactions on Intelligent Systems,2025,20(4):1046-1052.[doi:10.11992/tis.202506028]
点击复制
LI Deyi,LIU Yuchao,BAO Hong,et al.Understanding Albert Einstein: The “Four Elements Theory” is the first principle that governs human and machine cognition[J].CAAI Transactions on Intelligent Systems,2025,20(4):1046-1052.[doi:10.11992/tis.202506028]
读懂爱因斯坦——“四要素说”是统揽人和机器认知的第一性原理
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
20
期数:
2025年第4期
页码:
1046-1052
栏目:
认知物理学
出版日期:
2025-08-05
- Title:
- Understanding Albert Einstein: The “Four Elements Theory” is the first principle that governs human and machine cognition
- Keywords:
- first principle; Einstein; Four Elements Theory; cognitive physics; intelligence and energy relation function
- 分类号:
- TP18
- 文献标志码:
- 2025-7-4
- 摘要:
- 古希腊哲学家亚里士多德提出，可以用第一性原理回归到问题的本质来理解事物的规律。爱因斯坦在提出广义相对论之后，于20世纪20年代开始试图寻找一种更为宏大的统一理论来解释物质的相互作用，他试图用场论来作统一解释，但未能取得成功。本文指出，这种尝试需要区分物质世界的物理空间和人类精神世界里的认知空间，区分物质硬构体和思维软构体。宇宙是物质的，宇宙里不存在虚空，也不存在时间。虚空和时间是人类认知的精神产物，并非客观存在。人类对客观存在的大自然的认知是无尽的，不可能不带有渐进性、主观性和局限性，做不到绝对客观，尽管我们希望尽可能地客观。物质、能量、结构和时间构成人类认知和机器认知的最基础要素，“四要素说”是统揽人类认知和机器认知的第一性原理，通过“四要素说”可以更深入地理解事物之间的关联、生命的复杂过程，解释智能的产生，实现智能的度量，尤其是理解人的认知和机器认知的同源性和同一性。人工智能就是把生命体的认知从人的智慧中剥离出来，用时序维持负熵，用物理机器进行实际操作，通过机器的物质、结构变化和运行时隙精度的提高，用较低能耗实现甚至超过人的智能。
- Abstract:
- Aristotle, the ancient Greek philosopher, proposed that first principles could be used to return to the essence of a problem. After developing the general theory of relativity, Albert Einstein, in the 1920s, sought a more ambitious unified theory to explain the interaction of matter and complex physical phenomena using first principles. His attempts to use field theory as a unifying framework were ultimately unsuccessful. This paper argues that achieving such a unified theory necessitates distinguishing between the physical space of the natural world and the cognitive space of the human mental realm, as well as between the hard-structured ware of matter and the soft-structured ware of thought. The universe is fundamentally material; neither emptiness nor time exists objectively within it. Instead, space and time are spiritual constructs of human cognition, lacking independent, objective existence. Human cognition of objective nature is inherently endless and can never be absolutely objective, as it is inevitably progressive, subjective, and constrained. Nevertheless, striving for maximal objectivity remains a central aim. Matter, energy, structure, and time constitute the most fundamental elements of human and machine cognition. The “four elements theory” serves as the first principle to unify human and machine cognition. By employing this theory, we can better comprehend the complex processes of life, elucidate the origins and measurement of intelligence, and, importantly, understand the shared foundations and commonalities between human and machine cognition. Artificial intelligence is to make cognitive machines achieve, or even exceed, human intelligence with lower energy consumption through changes in machine structure and improvement in the accuracy of running time slots.
- 参考文献/References:
-
[1] 亚里士多德. 形而上学[M]. 苗力田, 译. 北京: 中国人民大学出版社, 2003: 12-14.
[2] 老子. 道德经[M]. 北京：群言出版社, 2014.
[3] 爱因斯坦. 爱因斯坦文集[M]. 许良英, 范岱年, 译. 北京: 商务印书馆, 1976: 558-559.
[4] MAXWELL J C. VIII. A dynamical theory of the electromagnetic field[J]. Philosophical transactions of the Royal Society of London, 1865(155): 459-512.
[5] 李德毅. 认知物理学基础: 认知自然和人类自身的奠基石[J]. 人民论坛·学术前沿, 2024, 294(14): 4-10.
[6] 李德毅, 淦文燕, 刘璐莹. 人工智能与认知物理学[C]//中国人工智能学会第10届全国学术年会. 广州：中国人工智能学会，2003.
[7] LI Deyi. Cognitive physics: the enlightenment by schr?dinger, turing, and wiener and beyond[J]. Intelligent Computing, 2023, 2: 0009.
[8] LI Deyi, DU Yi. Artificial intelligence with uncertainty[M]. Beijing: National Defense Industry Press. 2005.
[9] 李德毅, 殷嘉伦, 张天雷, 等. 机器认知四要素说[J]. 中国基础科学, 2023, 25(3): 1-10, 22.
LI Deyi, YIN Jialun, ZHANG Tianlei, et al. Four most basic elements in machine cognition[J]. China basic science, 2023, 25(3): 1-10, 22.
[10] SCHRODINGER E. What is life? the physical aspect of the living cell [J]. Cancer research, 1945, 5(11): 670–672.
[11] 李德毅. 人工智能看哲学[J]. 科学与社会, 2023, 13(2): 123-135.
LI Deyi. Artificial intelligence views philosophy[J]. Science and society, 2023, 13(2): 123-135.
[12] 毛泽东. 实践论[M]. 毛泽东. 毛泽东选集: 第一卷. 北京: 人民出版社出版, 1951.
[13] WIENER N. Cybernetics or control and communication in the animal and the machine[M]. Cambridge: The Technology Press, 1948.
[14] 毛泽东. 人的正确思想是从哪里来的?[M]. 北京: 人民出版社, 1964.
[15] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. Advances in neural information processing systems, 2017, 30.
[16] BIANCONI E, PIOVESAN A, FACCHIN F, et al. An estimation of the number of cells in the human body[J]. Annals of human biology, 2013, 40(6): 463-471.
[17] 李德毅. 论智能的困扰和释放[J]. 智能系统学报, 2024, 19(1): 249-257.
LI Deyi. On the puzzle and release of intelligence[J]. CAAI Transactions on Intelligent Systems, 2024, 19(1): 249-257.
[18] TURING A M. On computable numbers, with an application to the Entscheidungs problem[J]. Journal of mathematics, 1936, 58(345-363): 5.
[19] BROWN T B, MANN B, RYDER N, et al. Language models are few-shot learners[EB/OL]. (2020-05-28)[2025-06-25]. https://arxiv.org/abs/2005.14165.
[20] 弗伦奇 A. P. 牛顿力学[M]. 郭敦仁, 何成钧, 译. 北京: 人民教育出版社, 1978.
[21] 爱因斯坦. 相对论原理[M]. 赵志田, 刘一贯, 译. 北京: 科学出版社, 1980.
[22] HANOCH G, JüRGEN R. Relativity: the special and the general theory - 100th Anniversary Edition[M]. Princeton: Princeton University Press. 2015: 155-178.
[23] 毛泽东. 矛盾论[M]. 毛泽东. 毛泽东选集: 第一卷. 北京: 人民出版社出版, 1951.
备注/Memo
收稿日期:2025-6-25。
作者简介:李德毅，中国工程院院士，欧亚科学院院士，中国人工智能学会和中国指挥控制学会名誉理事长，清华大学博士生导师。主要研究方向为不确定性人工智能、数据挖掘、复杂网络、自驾驶和认知物理学。E-mail：lidy@cae.cn。;刘玉超，中国指挥与控制学会秘书处秘书长，主要研究方向为指挥系统技术和智能驾驶。E-mail：yuchao_liu@163.com。;鲍泓，二级教授，博士生导师，北京市信息服务工程重点实验室学术委员会副主任，中国人工智能学会智能驾驶专委会副主任，主要研究方向为视觉信息认知计算和智能驾驶。E-mail：baohong@buu.edu.cn。;蒋升，高级研究专员，Intelligent Computing编辑。主要研究方向为生物医学工程、复杂网络和认知物理学。E-mail：jiangsheng@zhejianglab.com。
通讯作者:蒋升，E-mail：jiangsheng@zhejianglab.com
更新日期/Last Update:
1900-01-01