Physicists have been pushing the boundaries of precision measurements with cutting-edge technology, but a new breakthrough using Artificial Intelligence (AI) has taken their research to unprecedented heights.
The Laser Interferometer Gravitational-Wave Observatory (LIGO), which detects gravitational waves passing through space-time, is the latest example. In 2015, LIGO detected its first gravitational wave, marking a significant milestone in the field of astrophysics. Since then, researchers have been working to improve its design and sensitivity.
In an effort to enhance the detector’s performance, Rana Adhikari, a physicist at the California Institute of Technology, turned to AI for assistance. He created a software suite called PyTheus, which uses machine learning algorithms to optimize experimental configurations. The tool enabled researchers to explore novel ideas and design more complex experiments.
One notable application of PyTheus is in entanglement swapping, a technique that creates quantum entanglement between particles with no shared past. Researchers used the AI-powered software to devise an experiment that successfully demonstrated this phenomenon, which has far-reaching implications for quantum technology.
AI is not only being used to design experiments but also to parse experimental results. For instance, Kyle Cranmer and his team used machine learning models to predict the density of dark matter clumps in the universe based on observable properties of nearby clumps. The AI’s equation provided a better fit to the data than human-made ones.
Rose Yu, a computer scientist at the University of California, San Diego, and her colleagues applied machine learning techniques to identify symmetries in data collected at the Large Hadron Collider. They discovered Lorentz symmetries, which are crucial to Einstein’s theories of relativity.
While AI has not yet led to new discoveries in physics, its potential is vast, according to experts like Aephraim Steinberg and Kyle Cranmer. As machine learning models become more sophisticated, they may aid researchers in creating hypotheses or explaining the underlying physics behind their findings.
The advent of large language models like ChatGPT could potentially bridge the gap between pattern discovery and hypothesis generation, according to Cranmer. This breakthrough has the potential to revolutionize physics research, enabling scientists to explore new frontiers and push the boundaries of human understanding.
Source: https://www.wired.com/story/ai-comes-up-with-bizarre-physics-experiments-but-they-work