Classical error correction methods are not suitable for quantum computing. The reason lies in quantum mechanics, which at a basic level does not allow recording intermediate results for further comparison. New error correction methods partially cope with this task, but have many limitations. Scientists from Austria were able to implement an error correction mechanism by connecting two different algorithms, thereby increasing the accuracy of calculations.
Image source: Helene Hainzer/University of Innsbruck
The intermediate quantum states of cubes involved in the calculations cannot, for example, can be preserved to verify parties. Therefore, one logical cubit is created from several physical cubes, while part of the physical cubes in the logical cubit are confused in a certain way. This allows you to track errors without destroying the calculation chain and adjust them.
The main difficulty lies in the fact that different groups of logical elements (gates) require different correction codes. Scientists from the University of Innsbruck (University of Innsbruck) have developed a technique that allows you to switch a quantum computer from one optimal code to another in the process of computing. This significantly reduced the frequency of errors.
The researchers tested their technique on a quantum computer with ion traps. The computer consisted of 16 qubits, from which two independent logic circuits were created. Each circuit was processed with an error correction code that was optimal for it. Switching between logical circuits occurred without errors, which confirmed the possibility of using two independent codes within one computational cycle.
In the future, this technique will simplify the correction of errors in the scaling of calculations, economically consuming physical cubes, which there will never be many.