AI Discovers Hidden Forces in Fourth State of Matter With 99% Accuracy
Scientists combine neural networks with dusty plasma experiments to reveal non-reciprocal forces and overturn long-held assumptions about particle interactions.
Physicists have achieved a major breakthrough in using artificial intelligence to uncover entirely new laws of nature, combining a specially designed neural network with precise 3D tracking of particles in dusty plasma to reveal hidden patterns in how particles interact. The research, published in PNAS by scientists at Emory University, focuses on the mysterious "fourth state of matter" found everywhere from space to wildfires, where ionized gas contains tiny charged dust grains that interact through complex forces.
The team's AI model captured non-reciprocal forces—where one particle influences another differently than it is influenced in return—with over 99 percent accuracy, a remarkable achievement given how difficult these forces are to measure and model. "We showed that we can use AI to discover new physics," says Justin Burton, an Emory professor of experimental physics and senior co-author of the paper. "Our AI method is not a black box: we understand how and why it works. The framework it provides is also universal. It could potentially be applied to other many-body systems to open new routes to discovery."
More significantly, the research overturned long-held assumptions about how these forces behave in complex systems. "What's even more interesting is that we show that some common theoretical assumptions about these forces are not quite accurate," explains Ilya Nemenman, an Emory professor of theoretical physics and co-senior author. "We're able to correct these inaccuracies because we can now see what's occurring in such exquisite detail." The ability to describe these forces with such precision opens new possibilities for understanding particle interactions across multiple fields.
The study offers one of the most detailed descriptions to date of the physics governing dusty plasma, a system that consists of ionized gas filled with interacting charged particles. Using their AI model, researchers were able to track particles with unprecedented precision using high-performance computing and machine-learning tools. The work was primarily supported by the National Science Foundation, with additional funding from the Simons Foundation, highlighting the collaborative nature of this interdisciplinary breakthrough.
The implications extend far beyond plasma physics, as the team believes this method could be applied broadly to systems made up of many interacting components. These range from industrial materials such as paint and ink to groups of living cells, potentially revolutionizing how scientists study complex systems. "This project serves as a great example of an interdisciplinary collaboration where the development of new knowledge in plasma physics and AI may lead to further advances in the study of living systems," says Vyacheslav Lukin, program director for the NSF Plasma Physics program. The research demonstrates that AI can move beyond data analysis to actively participate in scientific discovery.
