infinite-memory

[Noise-dominated regime] Suppose that there exist constantsξV >ξ U > 1,CU > 0 such thatUt =CUt−ξU (1 +o(1)) andVt =O(t−ξV ) ast→∞ · 2023

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

browse 1 citing papers

representative citing papers

math.OC · 2025-04-16 · conditional · novelty 7.0

Corner Gradient Descent achieves O(t^{-θζ}) loss rates for SGD on power-law spectral problems, with optimal θ = min(2, ν, 2/(ζ+1/ν)).

Showing 1 of 1 citing paper.

Corner Gradient Descent math.OC · 2025-04-16 · conditional · none · ref 4
Corner Gradient Descent achieves O(t^{-θζ}) loss rates for SGD on power-law spectral problems, with optimal θ = min(2, ν, 2/(ζ+1/ν)).