New Cold Atoms System Could Boost Quantum Capabilities

A new discovery could lead to artificial neural networks trapping cold atoms with high efficiency for use in quantum computers.

AI and quantum computing are two of the most disruptive technologies that, although still in their early days, can complement and accelerate one another.

Quantum systems can run machine-learning algorithms way more efficiently than conventional computers, — something that California-based startup Rigetti, for example, is working on.

AI can also unleash the power of quantum computers thanks to its ability to transcend human limits.

We’ve just reported about how the architecture of Google’s Go-playing AI AlphaGo has inspired scientists to build artificial neural networks to tackle the quantum decoherence issue and keep qubits stable.

Read More: How Neural Networks Cracked Quantum Error Correction

Now, another research team from the Australian National University (ANU) shows the potential of artificial deep neural networks in boosting the performance of quantum systems in another area.

AI To Optimize Input Parameters of Quantum Information

By cooling atoms to extremely low temperatures, physicists can produce qubits for quantum communication systems.

But cold atoms also provide the basis of ultraprecise devices that can measure time and space with unprecedented accuracy. Besides quantum computing, cold atoms find applications in telecommunications, navigation, and geophysics.

Recently, researchers at the ANU Department of Quantum Science have been working on trapping cold atoms to build quantum systems using artificial neural networks.

The end goal of the project is to build a quantum repeater, “a device that can be used to send quantum information over long distances. For that to work, we need to trap as many cold atoms in it as possible.”

Because trapping cold atoms is a delicate task that requires a lot of input and optimization, the ANU team developed an AI that can do that for them with greater efficiency.

“We use AI to control a large number of inputs to our experiment – the different laser and magnetic field settings – to seek out the best possible experimental conditions,” said Dr. Geoff Campbell, a post-doctoral fellow at the Centre for Quantum Computation and Communication Technology. “Because we have so many inputs we can only make educated guesses based on our understanding of what works best, but the AI is better at it than we are.”

According to researchers, deep neural networks can trap double the amount of cold atoms and cut this task time in half.

The artificial AI system the ANU team developed “found a solution that is highly effective and defies our intuition. One hundred students working for 100 years would probably never find it.”

The possibilities of this new discovery are just beginning to be realized. But, it is certain that it will help researchers in the quantum computing field to better understand the properties of sub-atomic particles.

Do you think quantum computers will become a reality in our lifetime?