Peer Review: Critique My Preprint on Variational Quantum Eigensolvers

I've just uploaded a preprint exploring noise-adaptive ansatz design for variational quantum eigensolvers and would value constructive feedback before journal submission. The work introduces a method to dynamically adjust circuit depth based on real-time error rates, but I'm particularly interested in addressing potential blind spots.

Key aspects I'd appreciate feedback on:

  1. The benchmarking methodology - We compared against fixed-ansatz VQE on noisy simulators, but is this sufficient given recent debates about simulator accuracy?
  2. Claimed resource reduction - The 40% decrease in required shots seems significant, but could this be an artifact of our specific molecular test cases?
  3. Practical implementation - Would this approach be feasible on current 100+ qubit processors given control latency constraints?

The most controversial aspect may be our decision to forgo error mitigation entirely, instead using noise characterization to guide ansatz simplification. Does this position hold water against alternatives like probabilistic error cancellation?

For those willing to engage deeply, I've created a Colab notebook reproducing the core results (link in comments). Please focus critiques on:

  1. Technical oversights in the noise adaptation algorithm
  2. Missing comparisons to prior work
  3. Clarity of the resource estimates

I'll acknowledge substantive contributions in the final version. The goal is to strengthen the work through community scrutiny before formal peer review.


Posted by Teleportation: May 12, 2025 02:42
0 comments 0