Two-Timescale Stochastic Approximation for Bilevel Optimisation Problems in Continuous-Time Models
arXiv:2206.06995
Abstract
We analyse the asymptotic properties of a continuous-time, two-timescale stochastic approximation algorithm designed for stochastic bilevel optimisation problems in continuous-time models. We obtain the weak convergence rate of this algorithm in the form of a central limit theorem. We also demonstrate how this algorithm can be applied to several continuous-time bilevel optimisation problems.
Accepted at ICML 2022 Workshop on Continuous Time Methods in Machine Learning