Pick up any physics textbook and you’ll find formula after formula describing how things wobble, fly, swerve, and stop. The formulas describe actions that we can observe, but behind each could be sets of factors that are not immediately obvious.
Now, a new AI program developed by researchers at Columbia University has apparently discovered its own alternative physics.
After seeing videos of physical phenomena on Earth, the AI did not rediscover the current variables we use; instead, he came up with new variables to explain what he saw.
To be clear, this does not mean that our current physics are faulty or that there is a model better suited to explain the world around us. (Einstein’s laws have proven to be incredibly robust.) But these laws could only exist because they were built on the back of a pre-existing “language” of theories and principles established by centuries of tradition. .
Given an alternate timeline where other minds approached the same issues from a slightly different perspective, would we still frame the mechanisms that explain our Universe in the same way?
Even with new technologies for imaging black holes and detecting strange and distant worlds, these laws have held up time and time again (additional note: quantum mechanics is a whole other story, but let’s stick to the visible world here) .
This new AI has only watched videos of a handful of physical phenomena, so it is in no way placed to come up with new physics to explain the Universe or try to better Einstein. That was not the goal here.
“I always wondered if we encountered an intelligent extraterrestrial race, would they have discovered the same physical laws as us, or could they describe the Universe in a different way?” says roboticist Hod Lipson of Columbia’s Creative Machines Lab.
“In the experiments, the number of variables was the same each time the AI was restarted, but the specific variables were different each time. So yes, there are alternative ways of describing the Universe and it is quite possible that our choices are not perfect.” .”
Beyond that, the team wanted to know if AI could actually find new variables – and therefore help us explain complex new phenomena emerging in our current deluge of data that we currently don’t have the theoretical understanding for. follow.
For example, new data from giant experiments such as the Large Hadron Collider that hint at new physics.
“What other laws are we missing just because we don’t have the variables?” says Columbia University mathematician Qiang Du.
So how does an AI find new physics? To begin, the team fed the system with raw video footage of phenomena they already understood and asked the program a simple question: what are the minimum fundamental variables needed to describe what is happening?
The first video showed a swinging double pendulum known to have four state variables in play: the angle and the angular velocity of each of the two pendulums.
The AI pondered the images and the question for a few hours, then spat out an answer: This phenomenon would require 4.7 variables to explain it, he said.
It’s pretty close to the four we know of…but it still didn’t explain what the AI thought the variables were.
So the team tried to match the known variables to the variables chosen by the AI. Two of them loosely corresponded to the angles of the arms, but the other two variables remained a mystery. Still, the AI could make accurate predictions about what the system would do next, so the team figured the AI must be onto something it couldn’t quite grasp.
“We tried to correlate the other variables with everything we could think of: angular and linear velocities, kinetic and potential energy, and various combinations of known quantities,” software researcher Boyuan Chen explains today. assistant professor at Duke University, who led the work.
“But nothing seemed to fit perfectly…we don’t yet understand the mathematical language he speaks.”
The team then showed more videos to the AI. The first featured an “air dancer” with a wavy arm blowing in the wind (the AI said this had eight variables). The lava lamp images also produced eight variables. A music video of flames returned with 24 variables.
Each time, the variables were unique.
“Without any prior knowledge of the underlying physics, our algorithm uncovers the intrinsic dimension of the observed dynamics and identifies candidate sets of state variables,” the researchers write in their paper.
This suggests that in the future, AI could potentially help us identify variables that underlie new concepts that we are not currently aware of. Watch this place.
The research has been published in Computational science of nature.