{"id":"arxiv/2604.00683","url":"https://arxiv.org/abs/2604.00683","tags":["stat.ME","math.OC","stat.ML"],"score":0,"title":"Convergence of projected stochastic natural gradient variational inference for various step size and sample or batch size schedules","domain":"stat.ME","source":"arxiv","authors":["Thomas Guilmeau","Hadrien Hendrikx","Florence Forbes"],"reviews":0,"subject":"stat.ME","summary":"Stochastic natural gradient variational inference (NGVI) is a popular and efficient algorithm for Bayesian inference. Despite empirical success, the convergence of this method is still not fully understood. In this work, we define and study a projected stochastic NGVI when variational distributions form an exponential family. Stochasticity arises when either gradients are intractable expectations or large sums. We prove new non-asymptotic convergence results for combinations of constant or decreasing step sizes and constant or increasing sample/batch sizes. When all hyperparameters are…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-04-01","reactions":{}}