Quantum computers have the potential to outperform even the most powerful supercomputers by leveraging the laws of quantum mechanics for their computation. The qubits (quantum bits) that perform those operations are in quantum states that are susceptible to environmental effects, which can lead to errors. For this reason, error correction is a big focus in the field. A company has now presented a bold new approach to how this can be done.
A quite common approach to error correction is to surround your computing qubit with a large number of qubits that can make sure it is functioning correctly. This number can vary between 1,000 and 10,000 qubits working to correct a single one. This is seen as a “brute force” approach to the problem, but it is clear that it requires a lot of qubits for not much computation.
Several models have tried to solve this, bringing the number of qubits down to at least a comparable number. Academic research has been looking at ways to work on correction directly on one qubit, with some breakthroughs in how long a qubit works before succumbing to the “noise” – temperature changes, magnetic fields, and a wide variety of other factors.
A single-qubit approach to correction has also now been announced by startup Nord Quantique. The Canadian company has announced that they can do it without extra qubits, the first in the industry to do so. Their method reduces errors by 14 percent on a single qubit by itself.
Their quantum system uses a cavity of ultrapure and superconducting aluminum. Microwave photons are introduced in the cavity and they are placed in specific quantum states to do the computation. The team believes that they can build a quantum computer with 100 logical qubits and a handful of physical ones that help reduce the other types of errors. Using photons as bosonic qubits they will be able to reach faster speeds, up to 1,000 times faster than other quantum computing architectures.
“Our model incorporates redundancy into every logical qubit, drastically reducing the number of physical qubits required for error correction once scaled. This positions us well to develop highly efficient and scalable quantum computers, without the need for vast amounts of physical qubits devoted to error correction, and potentially reaching fault-tolerance in a shorter time,” Julien Camirand Lemyre, President and CTO at Nord Quantique, said in a statement.
The road to fully scalable error-free quantum computers is far from straightforward, so the exploration of different methods is key to finding winning architectures that could move the technology forward.