Quantum Error Mitigation, Lecture 2

Part of the series Quantum Error Mitigation (Boulder Summer School 2023); cited as one entry below.

Explains why the inverse of a noise map is unphysical and how probabilistic error cancellation (PEC) implements it on average by sampling quantum trajectories with quasi-probabilities, using a random-walk analogy for intuition. It then covers measurement statistics and error bars, and how to learn device noise layer by layer: Pauli twirling, noise amplification, the degeneracy that keeps some Pauli fidelities of an entangling layer from being learned individually, and sparse Pauli-Lindblad noise tomograms up to 127 qubits.

• Canceling noise with noise: PEC on one qubit and the sampling overhead γ
• Bernoulli shot statistics and the empirical mean as an unbiased estimator
• Learning noise by amplification, and the degeneracy that limits learning entangling-gate noise
• Sparse Pauli-Lindblad model and noise tomograms for 20-qubit and 127-qubit Trotter layers

Summary and key points drafted by a language model from the slides, not yet reviewed by me. The slides are the source.

Cite

@misc{minev2023quantumerrormitigation,
  author = {Zlatko K. Minev},
  title = {{Quantum Error Mitigation (Boulder Summer School 2023)}},
  howpublished = {Lecture series, Boulder Summer School},
  year = {2023},
  url = {https://zlatko-minev.github.io/zlatko-minev-quantum-repository/series/quantum-error-mitigation-boulder-summer-school-2023/},
  note = {Slides, 7 decks}
}

Copy BibTeX

Reuse