Current error correction methods in quantum computers have improved 10 times: this is the important milestone achieved through a form of code by IBM researchers - led by Sergey Bravyi of the Watson Research Institute, and published in Nature - which opens the doors to the so-called 'era of utility', that of quantum computers truly capable of working on hitherto impossible problems.
"The main obstacle to the construction of quantum computers large enough to carry out advanced calculations is the fragility of quantum information compared to external noise sources", Fabio Sciarrino, of the Quantum Lab at Sapienza University in Rome, commented to ANSA. To solve the problem we have been working on several fronts for years, from the elimination of 'impurities' in materials to new methods for creating more 'copies' of qubits (the analogous basic unit of bits in traditional computers) to new algorithms. A sector that has seen great progress in the last year and now reaches a very important milestone that has earned the cover of Nature. Using new code, an algorithm to process the activities of qubits, the researchers managed to improve the correction of errors that accumulate on qubits by a factor of 10. "Using this type of code - added Federico Mattei, Quantum Business Developer at IBM - combined with a new hardware approach, in which it is possible to 'couple' distant qubits as if the chip had an extra dimension or was partly folded on itself, we have actually managed to reduce the number of qubits needed to perform the same operation by 10 times." An important progress that is part of IBM's roadmap to arrive at quantum processors with thousands of error-proof qubits in 4 or 5 years and which increasingly confirms the entry into the so-called 'era of utility'. A phase in which sufficient power is rapidly being reached for their concrete use, alongside traditional computers, for the study of extremely difficult concrete problems, and which marks the first steps towards the next objective: developing quantum computers capable of solving problems that are currently impossible even for the best supercomputers.
Reproduction reserved © Copyright ANSA