Conservative SPDEs as fluctuating mean field limits of stochastic gradient descent
Анотація
Abstract The convergence of stochastic interacting particle systems in the mean-field limit to solutions of conservative stochastic partial differential equations is established, with optimal rate of convergence. As a second main result, a quantitative central limit theorem for such SPDEs is derived, again, with optimal rate of convergence. The results apply, in particular, to the convergence in the mean-field scaling of stochastic gradient descent dynamics in overparametrized, shallow neural networks to solutions of SPDEs. It is shown that the inclusion of fluctuations in the limiting SPDE improves the rate of convergence, and retains information about the fluctuations of stochastic gradient descent in the continuum limit.
Класифікація
Ідентифікатори
Рецензії (0)
Написати рецензіюРецензій ще немає. Будьте першим!
Схожі роботи
Extended Poisson equation for weakly ergodic Markov processes
Схоже за: Stochastic processes and financial applications · Markov Chains and Monte Carlo Methods
Statistical Inference with Fractional Brownian Motion
Схоже за: Stochastic processes and financial applications · Statistical Mechanics and Entropy
Stochastic integral over the Arratia flow
Схоже за: Stochastic processes and financial applications · Markov Chains and Monte Carlo Methods
Divergence of an integral of a process with small ball estimate
Схоже за: Stochastic processes and financial applications · Statistical Mechanics and Entropy
Wick theorems in non-Gaussian white noise calculus
Схоже за: Stochastic processes and financial applications · Statistical Mechanics and Entropy
Poisson approximation of increment processes with Markov switching
Схоже за: Stochastic processes and financial applications · Markov Chains and Monte Carlo Methods