The day in conversation for 2026-11-16

discussion

Transcript

The discussion: Kai: I'm looking at the Grover search paper, specifically how they preserve those first peaks when noise is introduced; it seems like they’re proposing a threshold dependent on the specific noise model rather than a universal constant. Mira: That dependence is crucial. If we can nail down that exact dependency, does it change the complexity scaling compared to deterministic searches under similar noise constraints? Lev: It depends entirely on whether the underlying noise is correlated or purely depolarizing; I’m wondering if their framework implicitly assumes independent error channels, which might simplify things too much for real superconducting circuits. Kai: Exactly. If they assume independence, we need to know what the actual measurement fidelity looks like on a hardware implementation—is that fidelity high enough to support the proposed threshold, or are we operating in the regime where coherence times dominate? Mira: That brings up the LDPC code work; if we can’t reliably test that noise threshold mentioned above, how do we validate those error correction codes for superconducting hardware when the error profile isn't

More episodes

← Home