Imagine using artificial intelligence to compare two seemingly unrelated creations: biological tissue and Beethoven’s Ninth Symphony. At first glance, it may seem that living systems and musical masterpieces have nothing to do with each other. But a new AI method developed by Marcus J. Buehler, McAfee Professor of Engineering and Professor of Civil and Environmental Engineering and Mechanical Engineering at MIT, bridges this gap and reveals common patterns of complexity and order.
“By fusing generative AI with graph-based computational tools, this approach reveals completely new ideas, concepts, and designs that were previously unimaginable. , we can accelerate scientific discovery by teaching them to make novel predictions about design,” says Buehler.
This open-access research, recently published in Machine Learning: Science and Technology, demonstrates advanced AI techniques that integrate generative knowledge extraction, graph-based representation, and multimodal intelligent graph inference. Masu.
This study uses graphs, developed using methods inspired by category theory, as a central mechanism for teaching models to understand symbolic relationships in science. Category theory is a branch of mathematics that deals with abstract structures and the relationships between them, helping to understand and integrate diverse systems by focusing on objects and their interactions rather than their specific contents. provides a framework for In category theory, a system is viewed in terms of objects (which can be anything from numbers to more abstract entities such as structures or processes) and morphisms (arrows or functions that define relationships between these objects). Using this approach, Buehler was able to teach AI models to systematically reason about complex scientific concepts and behaviors. The symbolic relationships introduced through projection reveal that AI is not just drawing analogies, but engaging in deeper reasoning that maps abstract structures across different domains.
Buehler used this new method to analyze a collection of 1,000 scientific papers on biological materials and transform them into knowledge maps in the form of graphs. Graphs revealed how different pieces of information were connected, allowing us to find groups of related ideas and important points that connected many concepts.
“What’s really interesting is that graphs follow scale-free properties, are highly related, and can be effectively used for graph inference,” Buehler says. “In other words, we teach AI systems to think about graph-based data, building better world representation models and enhancing their ability to think and explore new ideas that enable discovery.”
Researchers use this framework to answer complex questions, find gaps in current knowledge, propose new designs for materials, predict how materials will behave, and You can connect unfamiliar concepts.
The AI model finds unexpected similarities between the biological material and the Ninth Symphony, suggesting that both follow a pattern of complexity. “Just as cells in biological materials interact in complex but organized ways to perform their functions, Beethoven’s Ninth Symphony arranges notes and themes to create a complex but coherent musical experience. ” says Buehler.
In another experiment, a graph-based AI model recommended creating a new biological material inspired by the abstract patterns seen in Wassily Kandinsky’s painting Composition VII. AI proposed a new mycelium-based composite material. “The result of this material combines a series of innovative concepts, including a balance between chaos and order, tunable properties, porosity, mechanical strength, and complex patterns of chemical functionality,” Buehler said. says. AI took inspiration from abstract paintings to create a material that is malleable and able to play a variety of roles, while maintaining a balance between strength and functionality. This application could lead to the development of innovative sustainable building materials, biodegradable alternatives to plastics, wearable technologies, and even biomedical devices.
Using this advanced AI model, scientists can extract insights from music, art, and technology, and analyze data from these fields to explore material design, research, and even the innovative potential of music and visual arts. You can identify hidden patterns that can cause the world.
“Graph-based generative AI enables much higher novelty than traditional approaches, exploring capabilities and technical details, and establishing a broadly useful framework for innovation by uncovering hidden connections. ” says Buehler. “This research not only contributes to the field of bio-inspired materials and mechanics, but interdisciplinary research leveraging AI and knowledge graphs will help the scientific community as we look to other studies in the future. It is also a stepping stone for the future, with the potential to become a tool for philosophical inquiry.”
