Sharper Bounds for Uniformly Stable Algorithms with Stationary Mixing Process

ICLR 2023 Motivation Generalization analysis of learning algorithms often builds on a critical assumption that training examples are independently identically distributed, which is often violated in practical problems such as time series prediction. A widely used relaxation of the i.i.d. assumption is to assume the observations are drawn from a mixing process, where the dependency between observations weakens over time. In this paper, we use algorithmic stability to study the generalization performance of learning algorithms with ψ-mixing data.