AI is Becoming Science's Best Friend
How machine learning can push the frontier of Particle Physics
August 24, 2018, Bruno Jacobsen
Science has, over the years, benefited tremendously from advances in computing power and other types of technology. Today, scientists, from chemists to cosmologists, can run more tests in a single day than their predecessors could in a year.
And while that's impressive, there's certainly more to come. AI, or ore specifically machine learning more, is not only making inroads into boardgames or self-driving cars. Besides the countless other fields it has already creeped into, AI is promising to revolutionize yet another one: physics.
In a recent article published in Nature, scientists have shown that AI is now at the frontier of physics' discoveries.
The standard model of particle physics has been the foundation of our understanding of much of nature thus far. But to dig further, we need experiments and we need to be able to interpret it. Some of these, for instance, are conducted in the Large Hadron Collider (LHC).
According to Manuel Gnida, the LHC, the particle accelerator at CERN, produces about a million gigabytes of data every second. And that's a lot. For that reason, much of this data is discarded because it simply cannot be interpreted.
However, advances in machine learning have turned AI into a tool that significantly increases the potential of single experiments for new discoveries and the furthering of our understanding of particle physics.
If you're interest in a robust breakdown of challenges and opportunities, you can find it in the Nature article. But what we find important to showcase how the number of fields AI is having a deep impact keeps growing.
And who knows - perhaps our next Einstein won't come in the form of a person, but 1s and 0s.