3 papers
cs.LG2023
Adam-like Algorithm with Smooth Clipping Attains Global Minima: Analysis Based on Ergodicity of Functional SDEs
Keisuke Suzuki
In this paper, we prove that an Adam-type algorithm with smooth clipping approaches the global minimizer of the regularized non-convex loss function. Adding smooth clipping and tak…
math.PR2022
Weak Convergence of Approximate reflection coupling and its Application to Non-convex Optimization
Keisuke Suzuki
In this paper, we propose a weak approximation of the reflection coupling (RC) for stochastic differential equations (SDEs), and prove it converges weakly to the desired coupling.…
math.PR2020
Construction and sample path properties of Brownian house-moving between two curves
Kensuke Ishitani, Daisuke Hatakenaka, Keisuke Suzuki
This study aims to construct a stochastic process called "Brownian house-moving," which is a Brownian bridge conditioned to stay between two curves. To construct this process, stat…