Abstract
AbstractWe study the mixing properties of an important optimization algorithm of machine learning: the stochastic gradient Langevin dynamics (SGLD) with a fixed step size. The data stream is not assumed to be independent hence the SGLD is not a Markov chain, merely a Markov chain in a random environment, which complicates the mathematical treatment considerably. We derive a strong law of large numbers and a functional central limit theorem for SGLD.
Funder
National Research, Development and Innovation Office
Publisher
Springer Science and Business Media LLC
Subject
Applied Mathematics,Control and Optimization
Reference17 articles.
1. Barkhagen, M., Chau, N.H., Moulines, É., Rásonyi, M., Sabanis, S., Zhang, Y.: On stochastic gradient Langevin dynamics with dependent data streams in the logconcave case. Bernoulli 27(1), 1–33 (2021)
2. Bradley, R.C.: Central limit theorems under weak dependence. J. Multivar. Anal. 11(1), 1–16 (1981)
3. Dey, S.: Online Learning: Sentiment Analysis with Logistic Regression Via Stochastic Gradient Ascent in Python. https://sandipanweb.wordpress.com/2017/03/31/online-learning-sentiment-analysis-with-logistic-regression-via-stochastic-gradient-ascent/
4. Gerencsér, B., Rásonyi, M.: On the ergodicity of certain Markov chains in random environments. J. Theor. Probab. 6, 1–33 (2023)
5. Gerencsér, B., Rásonyi, M.: Invariant Measures for Multidimensional Fractional Stochastic Volatility Models. Stochastics and Partial Differential Equations: Analysis and Computations, pp. 1–33, (2022)