(Super-)fast convergence rates for Chernoff approximations via centered differences and Richardson extrapolation
We use centered differences to construct Chernoff families and derive even asymptotic expansions for their iterated products. Richardson extrapolation is then used to cancel polynomial part and thereby raise the convergence order. As illustrations, we apply the construction to uniformly parabolic, Schrödinger-type second-order operators and to a third-order dispersive operator, obtaining, at extrapolation depth $m$, the rates $ O(n^{-2m}) $ on the Sobolev scales $H^{2m+δ}$, $H^{4m}$, and $H^{5m}$, respectively, with arbitrary $δ>0$. In the analytic category, quantitative all-order estimates…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2610.16094
curl -X POST /api/v1/records/arxiv/2610.16094/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2610.16094/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'