Introduces a class of asynchronous adaptive first-order methods and establishes O(1/sqrt t) convergence (up to logs) for non-convex stochastic optimization under reasonable assumptions.
LX ℓ=1 tr(log Πk,ℓ)− LX ℓ=1 tr(log Π−1,ℓ) # .(A.4) Then, for everyk≥0, ηE
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Stochastic convergence of parallel asynchronous adaptive first-order methods
Introduces a class of asynchronous adaptive first-order methods and establishes O(1/sqrt t) convergence (up to logs) for non-convex stochastic optimization under reasonable assumptions.