Math 2.0

Resources / preprint

Convex Optimization Is Free When Accuracy Is Expensive JHD3TW67QT

URIm2://res/JHD3TW67QT
Kindpreprint
CreatorsArthur Paing, Arthur Jacot
Date2026-09-28
MSC49-XX, 90-XX, 93-XX
Subjectsarxiv math.OC, arxiv cs.LG, arxiv stat.ML
Scopein

Expressions

e1: version, preprint

Languageen
Date2026-09-28
IdsarXiv 2609.35418

Items in this resource

The mentions of the inventory that name this resource. Each name links to the card of its inventory row.

ItemKindWhereSource
Convexstatementproposition 9arxiv
The step sizestatementcorollary 5arxiv
5TGY8Q2REYstatementproposition 11arxiv
Tilts are freestatementlemma 26arxiv
One stepstatementlemma 14arxiv
Strongly convexstatementproposition 10arxiv
Upgrading the oraclestatementproposition 3arxiv
The iterates stay in a ballstatementlemma 15arxiv
Every level coarser than $\delta$ is blindstatementlemma 25arxiv
Conversionstatementlemma 32arxiv
Noise floorstatementlemma 16arxiv
Transferstatementlemma 27arxiv
Cost-weighted hierarchy oracledefinitiondefinition 23arxiv
No speedup at the intrinsic ratestatementproposition 28arxiv
Optimalitystatementtheorem 7arxiv
Cost of optimizationstatementtheorem 4arxiv
Time-varying accuracystatementproposition 6arxiv
The residual is amortizedstatementlemma 33arxiv
No common approximate minimizerstatementlemma 24arxiv

Catalogue records