Efficient preconditioned stochastic gradient descent for estimation in latent variable models

被引:0
|
作者
Baey, Charlotte [1 ]
Delattre, Maud [2 ]
Kuhn, Estelle [2 ]
Leger, Jean-Benoist [3 ]
Lemler, Sarah [4 ]
机构
[1] Univ Lille, CNRS, UMR 8524 Labo Paul Painleve, F-59000 Lille, France
[2] Univ Paris Saclay, INRAE, MaIAGE, F-78350 Jouy En Josas, France
[3] Univ Technol Compiegne, CNRS, Heudiasyc, Compiegne, France
[4] Univ Paris Saclay, CentraleSupelec, Mathemat & Informat Complexite & Syst, F-91190 Gif Sur Yvette, France
关键词
MAXIMUM-LIKELIHOOD; INCOMPLETE DATA; APPROXIMATION; CONVERGENCE; OPTIMIZATION; ALGORITHM;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Latent variable models are powerful tools for modeling complex phenomena involving in particular partially observed data, unobserved variables or underlying complex unknown structures. Inference is often difficult due to the latent structure of the model. To deal with parameter estimation in the presence of latent variables, well-known efficient methods exist, such as gradient-based and EM-type algorithms, but with practical and theoretical limitations. In this paper, we propose as an alternative for parameter estimation an efficient preconditioned stochastic gradient algorithm. Our method includes a preconditioning step based on a positive definite Fisher information matrix estimate. We prove convergence results for the proposed algorithm under mild assumptions for very general latent variables models. We illustrate through relevant simulations the performance of the proposed methodology in a nonlinear mixed effects model and in a stochastic block model.
引用
收藏
页数:24
相关论文
共 50 条
  • [1] Preconditioned Stochastic Gradient Descent
    Li, Xi-Lin
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2018, 29 (05) : 1454 - 1466
  • [2] Stochastic Gradient Descent with Preconditioned Polyak Step-Size
    Abdukhakimov, F.
    Xiang, C.
    Kamzolov, D.
    Takac, M.
    [J]. COMPUTATIONAL MATHEMATICS AND MATHEMATICAL PHYSICS, 2024, 64 (04) : 621 - 634
  • [3] Preconditioned Stochastic Gradient Descent Optimisation for Monomodal Image Registration
    Klein, Stefan
    Staring, Marius
    Andersson, Patrik
    Pluim, Josien P. W.
    [J]. MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION (MICCAI 2011), PT II, 2011, 6892 : 549 - +
  • [4] Efficient estimation of generalized linear latent variable models
    Niku, Jenni
    Brooks, Wesley
    Herliansyah, Riki
    Hui, Francis K. C.
    Taskinen, Sara
    Warton, David I.
    [J]. PLOS ONE, 2019, 14 (05):
  • [5] Adjusted stochastic gradient descent for latent factor analysis
    Li, Qing
    Xiong, Diwen
    Shang, Mingsheng
    [J]. INFORMATION SCIENCES, 2022, 588 : 196 - 213
  • [6] Double Control Variates for Gradient Estimation in Discrete Latent Variable Models
    Titsias, Michalis K.
    Shi, Jiaxin
    [J]. INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 151, 2022, 151 : 6134 - 6151
  • [7] Uncertainty Sampling is Preconditioned Stochastic Gradient Descent on Zero-One Loss
    Mussmann, Stephen
    Liang, Percy
    [J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [8] On Estimation in Latent Variable Models
    Fang, Guanhua
    Li, Ping
    [J]. INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 139, 2021, 139
  • [9] A Variable Step Size Gradient-Descent TLS Algorithm for Efficient DOA Estimation
    Zhao, Haiquan
    Luo, Wenjing
    Liu, Yalin
    Wang, Chen
    [J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, 2022, 69 (12) : 5144 - 5148
  • [10] On Multilevel Monte Carlo Unbiased Gradient Estimation For Deep Latent Variable Models
    Shi, Yuyang
    Cornish, Rob
    [J]. 24TH INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS (AISTATS), 2021, 130