Quantum researchers advance error handling

0

A recent article published in Nature revealed a whole new phase of matter that has the potential to act as a long-term store of quantum information.

Researchers at the Center for Computational Quantum Physics at the Flatiron Institute in New York conducted an experiment that subjected qubits in a quantum computer to “quasi-rhythmic laser pulses” based on the Fibonacci sequence, demonstrating a way to store less error-prone quantum information. A Fibonacci sequence is a series of numbers where the next value in the sequence is calculated by adding the previous two numbers together (eg, 0, 1, 1, 2, 3, 5).

By projecting a sequence of laser pulses inspired by Fibonacci numbers at atoms inside a quantum computer, physicists created a new phase of matter that had never been observed before. Phase has the advantages of two time dimensions.

The researchers said the information stored in the phase is much more error-proof than with alternative configurations currently used in quantum computers. As a result, information can exist much longer without being scrambled – an important step in making quantum computing viable, said the study’s lead author Philipp Dumitrescu.

Dumitrescu led the theoretical part of the study with Andrew Potter from the University of British Columbia in Vancouver, Romain Vasseur from the University of Massachusetts in Amherst and Ajesh Kumar from the University of Texas at Austin. The experiments were performed on a quantum computer at Quantinuum in Broomfield, Colorado, by a team led by Brian Neyenhuis.

Quasicrystals

A typical crystal has a regular, repeating structure, like the hexagons in a honeycomb. A quasicrystal always has order, but its patterns never repeat. Quasicrystals are crystals of higher dimensions thrown or crushed into lower dimensions. These higher dimensions may even be beyond the three dimensions of physical space.

For qubits, Dumitrescu, Vasseur and Potter proposed in 2018 the creation of a quasicrystal in time, rather than space. While a periodic laser pulse would alternate (A, B, A, B, A, B, etc.), the researchers created a quasi-periodic laser pulse regime based on the Fibonacci sequence. In such a sequence, each part of the sequence is the sum of the previous two parts (A, AB, ABA, ABAAB, ABAABABA, etc.). This arrangement is ordered without repetition. It is also a 2D pattern crushed into one dimension.

The researchers tested the theory using Quantinuum’s quantum computer, pulsing laser light at the computer’s qubits both periodically and using the sequence based on Fibonacci numbers. The focus was on qubits at either end of the 10-atom range. Dumitrescu said: “With this quasi-periodic sequence, there is a complicated evolution that negates all the errors that live on the edge. Because of this, the edge remains quantum-mechanically consistent much, much longer than expected.

Towards error-free quantum computing

Meanwhile, in a recent blog post, IBM described its quantum error mitigation strategy as “the continuous path that will take us from the quantum hardware of today to the fault-tolerant quantum computers of tomorrow.”

In recent years, IBM said, its researchers have developed and implemented two general-purpose error mitigation methods, called zero-noise extrapolation (ZNE) and probabilistic error cancellation (PEC). The ZNE method cancels subsequent orders of noise affecting the expectation value of a noisy quantum circuit by extrapolating the measurement results to different noise intensities.

According to IBM, recent theoretical and experimental advances have shown that PEC can enable noiseless estimators of quantum circuits on noisy quantum computers. IBM predicted that its approach to error mitigation – which is analogous to how early classical computers developed – will allow it to develop quantum computers with more circuitry, which means greater power to solve problems. difficult problems.

One such difficult problem is weather forecasting, which involves the processing of complex nonlinear differential equations run on classical computer architectures.

Weather forecast

The recent heat wave across Europe has shown everyone the importance of accurate weather forecasts. BASF began exploring how the proprietary quantum algorithms developed by Pasqal could one day be used to predict weather conditions to support its digital agriculture business. Using parameters generated by weather models, BASF will be able to simulate crop yields and growth stages, as well as predict drift when applying crop protection products.

Advanced weather and climate modeling is typically run on conventional computers using Physics-Informed Neutral Networks (PINNs). According to Hyperion Research, 5% of global investment in high-performance computing (HPC) is focused on weather modeling.

Rather than relying on HPC, Pasqal aims to solve the underlying complex nonlinear differential equations in a way he calls “new and more efficient” by implementing quantum neural networks on his quantum processors at neutral atoms.

John Manobianco, Senior Weather Modeler in BASF’s Agricultural Solutions Division, said, “Leveraging Pasqal’s innovation for weather modeling validates the ability of quantum computing to go beyond what can be achieved with conventional high-performance computing. Such transformational technology can help us prepare for the impacts of climate change and move towards a more sustainable future.

These algorithms will only be viable when researchers and quantum computing companies improve error handling. However, some of the techniques used to solve problems can be executed today on conventional computer architectures.

For example, in a recent podcast, Bloomberg CTO Shawn Edwards explained why he thinks mainstream quantum computing is many years away. While a lot of progress has been made on the science behind it, Edwards said some of the most useful things to come out of quantum computing are quantum computing-inspired algorithms. He said Bloomberg’s quantum teams were looking to improve some algorithms based on quantum computing.

Such quantum-inspired algorithms may be the bridge that enables mass adoption of quantum computing. Even though error correction is still years away, research to improve error handling and the development of quantum-inspired code can encourage more IT managers to plan ahead and develop an IT strategy that incorporates quantum computing.

Share.

About Author

Comments are closed.