Quantum Error Correction and Generalized Measurements (POVMs)

Stub. Protecting quantum information from the noisy channels of the previous note via quantum error-correcting codes (encoding into a redundant subspace so errors are detectable/correctable without directly measuring, and hence destroying, the encoded state); and positive operator-valued measures (POVMs) and quantum instruments as the most general notion of a quantum measurement, more general than the projective measurements covered earlier.

· 1 min read · 170 words

Corresponds to Lectures 20–21. Quantum error correction faces a constraint classical error correction doesn’t: you can’t just measure the encoded data to check for errors, since measurement generally destroys quantum information. The resolution is to encode into a carefully chosen subspace such that error syndromes can be measured (revealing which error occurred) without revealing anything about the encoded logical state itself, allowing correction. POVMs generalize the measurement axiom of quantum-measurement-and-steering: instead of requiring orthogonal projectors, a POVM is any set of positive operators summing to the identity, which is exactly what arises when a projective measurement is performed on a larger system (via Stinespring) and only the outcome statistics on the original system are kept — the most general measurement quantum mechanics allows, with quantum instruments further tracking the post-measurement state, not just the outcome probabilities.

Outline (TODO — flesh out each)

  • Quantum error correction — encoding, syndromes, correction
  • Example codes
  • Positive operator-valued measures (POVMs)
  • Quantum instruments

← All notes