3 papers
math.OC2026
Randomized Subspace Nesterov Accelerated Gradient
Gaku Omiya, Pierre-Louis Poirion, Akiko Takeda
Randomized-subspace methods reduce the cost of first-order optimization by using only low-dimensional projected-gradient information, a feature that is attractive in forward-mode a…
math.OC2026
Convergence Analysis of Randomized Subspace Normalized SGD under Heavy-Tailed Noise
Gaku Omiya, Pierre-Louis Poirion, Akiko Takeda
Randomized subspace methods reduce per-iteration cost; however, in nonconvex optimization, most analyses are expectation-based, and high-probability bounds remain scarce even under…
stat.CO2025
Minimization of Functions on Dually Flat Spaces Using Geodesic Descent Based on Dual Connections
Gaku Omiya, Fumiyasu Komaki
We propose geodesic-based optimization methods on dually flat spaces, where the geometric structure of the parameter manifold is closely related to the form of the objective functi…