Introduction Artificial intelligence (AI) has evolved along two distinct paths: symbolic AI and numeric AI. In this blog post, we explore a novel approach to understanding transformer models, which are primarily numeric and data-driven, through the lens of Feynman diagrams, a symbolic representation used in quantum field theory. Our goal is to bridge the […]
Category Archives: Data Science
A Feynman Diagram-Inspired Approach to Understanding Transformer Models Abstract: The divide between symbolic AI and numeric AI has led to a range of AI models and techniques that rely on either formal, rule-based systems or data-driven, neural network-based approaches. This paper aims to explore the possibility of bridging this gap by drawing inspiration from Feynman […]
Visualizing AI Interactions: A Feynman Diagram-Inspired Approach to Understanding Transformer Models Abstract: In this paper, we propose an analogy between the input path of transformer AI models and Feynman diagrams, a well-established graphical representation of particle interactions in quantum field theory. While recognizing the fundamental differences between these two domains, our goal is to explore […]