Study suggests that physical processes can have hidden neural network-like abilities

https://phys.org/news/2024-01-physical-hidden-neural-network-abilities.html

By the University of Chicago

A new study, led by a UChicago graduate student in physics, has revealed that physical processes may harbour hidden neural network-like capabilities. This discovery challenges traditional scientific understanding and has far-reaching implications for technology and research. The study’s findings, published on Phys.org, shed light on the intricate relationship between physical phenomena and neural networks, opening up new possibilities for innovation and exploration.