Finite Sample and Large Deviations Analysis of Stochastic Gradient Algorithm with Correlated Noise
arXiv:2410.08449v1h-index: 1
Originality Synthesis-oriented
AI Analysis
This work provides theoretical guarantees for stochastic gradient algorithms in scenarios with correlated noise, which is incremental as it extends existing analysis to more realistic settings.
The paper analyzes the finite sample regret of a stochastic gradient algorithm with decreasing step size under correlated noise conditions, using a perturbed Lyapunov function for analysis and examining escape times via large deviations theory.
We analyze the finite sample regret of a decreasing step size stochastic gradient algorithm. We assume correlated noise and use a perturbed Lyapunov function as a systematic approach for the analysis. Finally we analyze the escape time of the iterates using large deviations theory.