基于一致遍历马氏链样本的分布式Huber回归算法的泛化性OA

This paper investigates the generalization performance of distributed Huber regression with samples generated from uniformly ergodic Markov chains.We first construct a distributed Huber regres-sion algorithm applicable to uniformly ergodic Markov chain samples.Relying on statistical learning the-ory,we further derive the convergence rates and generalization bounds for the proposed distributed Hu-ber regression with Markov chain sampling.Theoretical proofs demonstrate that the distributed Huber regression algorithm achieves a fast convergence rate even in the non-independent and identically distrib-uted(non-i.i.d.)sample setting.