How Quantum Computers Could Kill the Arrow of Time

A new technique for quantum computing could bust open our whole model of how time moves in the universe.

Here's what's long seemed to be true: Time works in one direction. The other direction? Not so much.

That's true in life. (Tuesday rolls into Wednesday, 2018 into 2019, youth into old age.) And it's true in a classical computer. What does that mean? It's much easier for a bit of software running on your laptop to predict how a complex system will move and develop in the future than it is to recreate its past. A property of the universe that theorists call "causal asymmetry" demands that it takes much more information — and much more complex calculations — to move in one direction through time than it does to move in the other. (Practically speaking, going forward in time is easier.)

This has real-life consequences. Meteorologists can do a reasonably good job of predicting whether it will rain in five days based on today's weather radar data. But ask the same meteorologists to figure out whether it rained five days ago using today's radar images? That's a much more challenging task, requiring a lot more data and much bigger computers. [The 18 Biggest Unsolved Mysteries in Physics]

Information theorists suspected for a long time that causal asymmetry might be a fundamental feature of the universe. As long ago as 1927, the physicist Arthur Eddington argued that this asymmetry is the reason we only move forward through time, and never backward. If you understand the universe as a giant computer constantly calculating its way through time, it's always easier — less resource-intensive — for things to flow forward (cause, then effect) than backward (effect, then cause). This idea is called the "arrow of time."

But a new paper, published July 18 in the journal Physical Review X, opens the door to the possibility that that arrow is an artifact of classical-style computation — something that's only appeared to us to be the case because of our limited tools.

A team of researchers found that in certain circumstances causal asymmetry disappears inside quantum computers, which calculate in an entirely different way— Unlike classical computers in which information is stored in one of two states (1 or 0), with quantum computers, information is stored in subatomic particles that follow some bizarre rules and so can each can be in more than one state at the same time. And, even more enticingly, their paper points the way toward future research that could show causal asymmetry doesn't really exist in the universe at all.

How's that?

Very orderly and very random systems are easy to predict. (Think of a pendulum — ordered — or a cloud of gas filling a room — disordered.) In this paper, the researchers looked at physical systems that had a goldilocks' level of disorder and randomness — not too little, and not too much. (So, something like a developing weather system.) These are very difficult for computers to understand, said study co-author Jayne Thompson, a complexity theorist and physicist studying quantum information at the National University of Singapore. [Wacky Physics: The Coolest Little Particles in Nature]

Next, they tried to figure out those systems' pasts and futures using theoretical quantum computers (no physical computers involved). Not only did these models of quantum computers use less memory than the classical computer models, she said, they were able to run in either direction through time without using up extra memory. In other words, the quantum modelshad no causal asymmetry.

"While classically, it might be impossible for the process to go in one of the directions [through time]," Thompson told Live Science, "our results show that 'quantum mechanically,' the process can go in either direction using very little memory."

And if that's true inside a quantum computer, that's true in the universe, she said.

Quantum physics is the study of the strange probabilistic behaviors of very small particles — all the very small particles in the universe. And if quantum physics is true for all the pieces that make up the universe, it's true for the universe itself, even if some of its weirder effects aren't always obvious to us. So if a quantum computer can operate without causal asymmetry, then so can the universe.

Of course, seeing a series of proofs about how quantum computers will one day work isn't the same thing as seeing the effect in the real world. But we're still a long way off from quantum computers advanced enough to run the kind of models this paper describes, they said.

What's more, Thompson said, this research doesn't prove that there isn't any causal asymmetry anywhere in the universe. She and her colleagues showed there is no asymmetry in a handful of systems. But it's possible, she said, that there are some very bare-bones quantum models where some causal asymmetry emerges.

"I'm agnostic on that point," she said.

For now.

The next step for this research, she said, is to answer that question — to figure out whether causal asymmetry exists in any quantum models.

This paper doesn't prove that time doesn’t exist, or that we’ll one day be able to slip backward through it. But it does appear to show that one of the key building blocks of our understanding of time, cause and effect, doesn't always work in the way scientists have long assumed — and might not work that way at all. What that means for the shape of time, and for the rest of us, is still something of an open question.

The real practical benefit of this work, she said, is that way down the road quantum computers might be capable of easily running simulations of things (like the weather) in either direction through time, without serious difficulty. That would be a sea change from the current classical-modeling world.

Can these really be true? please share your thoughts this.