New quantum mean estimators and SGD variants achieve query complexity Õ(√d ε^{-(5p-4)/(2p-2)}) for nonconvex and Õ(√d ε^{-(3p-2)/(2p-2)} + ε^{-2}) for convex heavy-tailed stochastic optimization, improving on classical lower bounds in low dimension.
Sublinear classical and quantum algorithms for general matrix games.Proceedings of the AAAI Conference on Artificial Intelligence, 35 (10):8465–8473, 2021
1 Pith paper cite this work, alongside 5 external citations. Polarity classification is still indexing.
1
Pith paper citing it
5
external citations · OpenAlex
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Quantum Speedups for Stochastic Optimization with Heavy-Tailed Noise
New quantum mean estimators and SGD variants achieve query complexity Õ(√d ε^{-(5p-4)/(2p-2)}) for nonconvex and Õ(√d ε^{-(3p-2)/(2p-2)} + ε^{-2}) for convex heavy-tailed stochastic optimization, improving on classical lower bounds in low dimension.