FROM THE YIJING TO NEURAL NETWORKS: A MATHEMATICAL COMPARISON OF ANCIENT DIVINATORY LOGIC AND MODERN MACHINE LEARNING ARCHITECTURES

Abstract

This paper presents a comparative mathematical analysis between the ancient divinatory logic of the Yijing (I Ching) and modern machine learning architectures. We demonstrate that the Yijing's 64-hexagram system, grounded in binary opposition (Yin-Yang), constitutes a formal computational framework that parallels foundational concepts in contemporary artificial intelligence. Through systematic examination of structural correspondences including binary encoding, probabilistic reasoning, and pattern recognition we establish that both systems operate on similar mathematical principles despite their temporal and cultural separation. Our methodology involves formalizing the Yijing's divination process as a computational model and comparing its architectural features with neural network paradigms. The results reveal significant structural isomorphisms between hexagram transformations and neural activation patterns, suggesting that ancient divinatory logic embodies fundamental computational principles that continue to inform modern machine learning. This cross-temporal analysis contributes to both the historical understanding of computational thought and the philosophical foundations of artificial intelligence.