Approximate blocked Gibbs sampling for Bayesian neural networks

被引:0
|
作者
Theodore Papamarkou
机构
[1] The University of Manchester,Department of Mathematics
来源
Statistics and Computing | 2023年 / 33卷
关键词
Approximate MCMC; Bayesian inference; Bayesian neural networks; Blocked Gibbs sampling; Minibatch sampling; Posterior predictive distribution;
D O I
暂无
中图分类号
学科分类号
摘要
In this work, minibatch MCMC sampling for feedforward neural networks is made more feasible. To this end, it is proposed to sample subgroups of parameters via a blocked Gibbs sampling scheme. By partitioning the parameter space, sampling is possible irrespective of layer width. It is also possible to alleviate vanishing acceptance rates for increasing depth by reducing the proposal variance in deeper layers. Increasing the length of a non-convergent chain increases the predictive accuracy in classification tasks, so avoiding vanishing acceptance rates and consequently enabling longer chain runs have practical benefits. Moreover, non-convergent chain realizations aid in the quantification of predictive uncertainty. An open problem is how to perform minibatch MCMC sampling for feedforward neural networks in the presence of augmented data.
引用
收藏
相关论文
共 50 条
  • [1] Approximate blocked Gibbs sampling for Bayesian neural networks
    Papamarkou, Theodore
    STATISTICS AND COMPUTING, 2023, 33 (05)
  • [2] GIBBS SAMPLING IN BAYESIAN NETWORKS
    HRYCEJ, T
    ARTIFICIAL INTELLIGENCE, 1990, 46 (03) : 351 - 363
  • [3] Gibbs sampling the posterior of neural networks
    Piccioli, Giovanni
    Troiani, Emanuele
    Zdeborova, Lenka
    JOURNAL OF PHYSICS A-MATHEMATICAL AND THEORETICAL, 2024, 57 (12)
  • [4] Estimating bayesian networks parameters using EM and Gibbs sampling
    Chai, Huimin
    Lei, Jiangnan
    Fang, Min
    8TH INTERNATIONAL CONFERENCE ON ADVANCES IN INFORMATION TECHNOLOGY, 2017, 111 : 160 - 166
  • [5] Approximate Bayesian neural networks in genomic prediction
    Waldmann, Patrik
    GENETICS SELECTION EVOLUTION, 2018, 50
  • [6] On the Expressiveness of Approximate Inference in Bayesian Neural Networks
    Foong, Andrew Y. K.
    Burt, David R.
    Li, Yingzhen
    Turner, Richard E.
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
  • [7] Approximate Bayesian neural networks in genomic prediction
    Patrik Waldmann
    Genetics Selection Evolution, 50
  • [8] Multilocus linkage analysis by blocked Gibbs sampling
    Thomas, A
    Gutin, A
    Abkevich, V
    Bansal, A
    STATISTICS AND COMPUTING, 2000, 10 (03) : 259 - 269
  • [9] Multilocus linkage analysis by blocked Gibbs sampling
    Alun Thomas
    Alexander Gutin
    Victor Abkevich
    Aruna Bansal
    Statistics and Computing, 2000, 10 : 259 - 269
  • [10] On MCMC sampling in Bayesian MLP neural networks
    Vehtari, A
    Särkkä, S
    Lampinen, J
    IJCNN 2000: PROCEEDINGS OF THE IEEE-INNS-ENNS INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOL I, 2000, : 317 - 322