Certification failure in variational-principle observables
Algorithms which prepare states via the variational principle assume that closeness in energy implies closeness in state; we formalise how this assumption can fail by showing that a ground-state energy estimate converged to within $ε$ of the true value is compatible with observable expectation values in error by up to $\sqrt{2εχ_O}$ at leading order, where $χ_O$ is the static susceptibility of the observable. As a result, the observable error can be orders of magnitude larger than the energy error, a mechanism we call non-linear error amplification, and because a variational algorithm cannot…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2610.03989
curl -X POST /api/v1/records/arxiv/2610.03989/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2610.03989/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'