Sample-efficient neural likelihood-free Bayesian inference of implicit HMMs

被引:0
|
作者
Ghosh, Sanmitra [1 ,2 ]
Birre, Paul J. [2 ,3 ]
De Angelis, Daniela [2 ,3 ]
机构
[1] PhysicsX, London, England
[2] Univ Cambridge, MRC Biostat Unit, Cambridge, England
[3] UK Hlth Secur Agcy, London, England
基金
英国医学研究理事会;
关键词
COMPUTATION; SIMULATION;
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
Likelihood-free inference methods based on neural conditional density estimation were shown to drastically reduce the simulation burden in comparison to classical methods such as ABC. When applied in the context of any latent variable model, such as a Hidden Markov model (HMM), these methods are designed to only estimate the parameters, rather than the joint distribution of the parameters and the hidden states. Naive application of these methods to a HMM, ignoring the inference of this joint posterior distribution, will thus produce an inaccurate estimate of the posterior predictive distribution, in turn hampering the assessment of goodness-of-fit. To rectify this problem, we propose a novel, sample-efficient likelihood-free method for estimating the high-dimensional hidden states of an implicit HMM. Our approach relies on learning directly the intractable posterior distribution of the hidden states, using an autoregressive-flow, by exploiting the Markov property. Upon evaluating our approach on some implicit HMMs, we found that the quality of the estimates retrieved using our method is comparable to what can be achieved using a much more computationally expensive SMC algorithm.
引用
收藏
页数:24
相关论文
共 50 条
  • [41] COSMOABC: Likelihood-free inference via Population Monte Carlo Approximate Bayesian Computation
    Ishida, E. E. O.
    Vitenti, S. D. P.
    Penna-Lima, M.
    Cisewski, J.
    de Souza, R. S.
    Trindade, A. M. M.
    Cameron, E.
    Busti, V. C.
    ASTRONOMY AND COMPUTING, 2015, 13 : 1 - 11
  • [42] A General Recipe for Likelihood-free Bayesian Optimization
    Song, Jiaming
    Yu, Lantao
    Neiswanger, Willie
    Ermon, Stefano
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 162, 2022,
  • [43] Likelihood-Free Bayesian Analysis of Memory Models
    Turner, Brandon M.
    Dennis, Simon
    Van Zandt, Trisha
    PSYCHOLOGICAL REVIEW, 2013, 120 (03) : 667 - 678
  • [44] Likelihood-free Bayesian inference framework for sizing kiwifruit from orchard imaging surveys
    Ranjard, Louis
    Bristow, James
    Hossain, Zulfikar
    Orsi, Alvaro
    Kirkwood, Henry J.
    Gilman, Andrew
    PROCEEDINGS OF THE 2021 36TH INTERNATIONAL CONFERENCE ON IMAGE AND VISION COMPUTING NEW ZEALAND (IVCNZ), 2021,
  • [45] Towards end-to-end likelihood-free inference with convolutional neural networks
    Radev, Stefan T.
    Mertens, Ulf K.
    Voss, Andreas
    Koethe, Ullrich
    BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY, 2020, 73 (01): : 23 - 43
  • [46] Detecting conflicting summary statistics in likelihood-free inference
    Yinan Mao
    Xueou Wang
    David J. Nott
    Michael Evans
    Statistics and Computing, 2021, 31
  • [47] Detecting conflicting summary statistics in likelihood-free inference
    Mao, Yinan
    Wang, Xueou
    Nott, David J.
    Evans, Michael
    STATISTICS AND COMPUTING, 2021, 31 (06)
  • [48] Bayesian model updating of civil structures with likelihood-free inference approach and response reconstruction technique
    Ni, Pinghe
    Han, Qiang
    Du, Xiuli
    Cheng, Xiaowei
    MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2022, 164
  • [49] Likelihood-Free Inference in High-Dimensional Models
    Kousathanas, Athanasios
    Leuenberger, Christoph
    Helfer, Jonas
    Quinodoz, Mathieu
    Foll, Matthieu
    Wegmann, Daniel
    GENETICS, 2016, 203 (02) : 893 - +
  • [50] Likelihood-free inference with neural compression of DES SV weak lensing map statistics
    Jeffrey, Niall
    Alsing, Justin
    Lanusse, Francois
    MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY, 2021, 501 (01) : 954 - 969