Two-timescale learning algorithms converge with realistic noise conditions

Original: Convergence of Two-Timescale Markovian Stochastic Approximations with Applications in Reinforcement Learning

Source: arxiv.org ↗