What is quantum error correction?
Methods that protect quantum information by spreading it over many qubits and repeatedly checking for errors.
Ordinary computers can guard against errors by keeping copies and taking a vote. Qubits can’t be copied (no-cloning), and measuring them destroys their state. Error correction gets around both by measuring only how groups of qubits relate to each other, such as “do these two agree?”, which reveals errors without revealing the data.
The method only helps if the physical qubits are already good enough. Below that level, adding qubits removes errors faster than it adds them. In 2024, Google’s Willow chip showed this happening: each step up in code size roughly halved the error rate.
Related
- Logical qubitA reliable qubit built from many physical qubits using error correction.
- Physical qubitOne actual qubit on a chip or in a trap. Physical qubits make errors, roughly once every few hundred to few thousand operations today.
- DecoherenceThe loss of a qubit's blend through contact with its surroundings, which turns it into ordinary randomness.
- No-cloning theoremThe rule that an unknown quantum state can't be copied.