A team of researchers in the United States at Purdue University of West Lafayette, Indiana, in conjunction with experts at Tohoku University in Sendai, Japan, are on the brink of developing the next generation of computers. This next generation is made out of p-bits to build technology that is “naturally probabilistic”.
What’s a probabilistic computer?
To fully grasp the profundity of this new technology, a few definitions and explanations are in order. Simply put, today’s computers, from PCs to tablets to smartphones, run on “building blocks called bits,” little pieces of information. These bits, according to Purdue, “are deterministically 0 or 1.” It’s very much like a black and white situation, a yes or no. “At the other end of the spectrum are quantum computers consisting of qubits which occupy some superposition of 0 and 1.” These qubits are the key to the ever legendary quantum tech that is being whispered about lately, very complex and not well understood.
However, “the p-bit fluctuates probabilistically between 0 and 1 and can be pinned to one or the other state based on the magnitude and sign of its input,” according to Purdue University. While these aren’t quite as complex as qubits, “there is a useful subset of problems solvable with qubits that can also be solved with p-bits,” said Supiyo Datta, Purdue’s Thomas Duncan Distinguished Professor of Electrical and Computer Engineering. He simplifies this idea further, explaining that “You might say that a p-bit is a poor man’s qubit.”
Can engineers build a computer able to solve quantum problems, without a quantum computer?#PurdueECE researchers – led by Prof. Supriyo Datta – and a team from Tohoku University are demonstrating how a probabilistic computer would do just that.
Details: https://t.co/OLx3ykBzO1 pic.twitter.com/MOsiMsPrQu
— PurdueECE (@PurdueECE) September 19, 2019
Now, how does that apply to this new “probabilistic” technology? A probabilistic computer, using p-bits, can more closely act like nature, constantly shifting and changing. This also means that it “may sooner solve some of the problems a quantum computer would solve since it wouldn’t need entirely new hardware or extremely cold temperatures to operate,” according to Purdue.
With the collaboration of these two schools, Purdue University and Tohoku University designed this computer. This was first demonstrated in 2019, aiming to tackle problems typically associated with quantum computing, find the nature of objects and problems, and sorting through a myriad of variables such as “drug research, encryption, cybersecurity, financial services, data analysis, and supply chain logistics”.
Why are these computers needed?
To make this concept clearer, the researchers used an average cup of coffee as an example. This is because it is something that has yet to be identified by quantum computers as these are being developed by Google and IBM. Purdue University explains that “The molecular structure of caffeine is so complex that classical computers can’t perform the calculations needed to understand it fully. This is because caffeine can exist in 10^48 different atomic configurations, or ‘quantum states.'”
Purdue University expounds further that “A classical computer, which processes only one quantum state at a time, would need to process many states at once like nature does to capture caffeine.” Thus the need for probabilistic computers.
Researchers from @PurdueEngineers are creating a new kind of computing that's "naturally probabilistic" – to bridge the gap between classical & quantum computing to more efficiently solve problems.
FIND OUT HOW: https://t.co/287WCV6rui #Purdue #PRF #Innovation #cybersecurity pic.twitter.com/L9Rl6h7QY1
— Purdue OTC (@PurdueOTC) March 10, 2021
The future is probably probabilistic
The whole mind-boggling project is still a work in progress, as all great undertakings always are. However, Physicist Richard Feynman explained in 1981 before a computer science conference that these probabilistic computers don’t need to be perfect to perform their job successfully. He said that “We could imagine and be perfectly happy, I think, with a probabilistic simulator of a probabilistic nature, in which the machine doesn’t exactly do what nature does, but…you’d get the corresponding probability with the corresponding accuracy.”
However, until we can truly see the implications of such a mysterious technology, the initiative responsible for this project, known as Purdue-P, will continue development on building towards the future, bit by bit. “As a field, we look at the computing problems we can’t solve yet and think,” said former Purdue Postdoctoral Researcher Karem Camsari. “There’s digital computing, there’s quantum computing – what else is there?”
Photo credit: The featured image has been illustrated by Gwen Keraval for Purdue University and has been provided for press usage.
