Bayesian model comparison versus generalization ability of neural networks

被引：0
|
作者：
Gomari, M [1 ]
Järvi, T [1 ]
机构：
[1] Turku Ctr Comp Sci, FIN-20520 Turku, Finland
来源：
INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL I AND II | 1999年
关键词：
neural networks; classification; Bayesian inference; evidence; generalization;
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：
Generalization ability is a desired feature for any model used for predicting new situation based on some previously learned knowledge. Often a neural network model with a good performance on the training cases does not provide an adequate performance for the unseen cases, i.e. the model is called to have a poor generalization ability. In this paper we discuss the applicability of the Bayesian techniques for measuring the generalization ability of neural networks through a medical case study.
引用
收藏
页码：537 / 541
页数：5
相关论文
共 50 条
  • [41] GENERALIZATION ABILITY OF MOS PREDICTION NETWORKS
    Cooper, Erica
    Huang, Wen-Chin
    Toda, Tomoki
    Yamagishi, Junichi
    2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2022, : 8442 - 8446
  • [42] Generalization theory and generalization methods for neural networks
    Wei, Hai-Kun
    Xu, Si-Xin
    Song, Wen-Zhong
    Zidonghua Xuebao/Acta Automatica Sinica, 2001, 27 (06): : 806 - 815
  • [43] Predicting indoor particle concentration in mechanically ventilated classrooms using neural networks: Model development and generalization ability analysis
    Ren, Jianlin
    He, Junjie
    Novoselac, Atila
    BUILDING AND ENVIRONMENT, 2023, 238
  • [44] Localized generalization error model for Multilayer Perceptron Neural Networks
    Yang, Fei
    Ng, Wing W. Y.
    Tsang, Eric C. C.
    Zeng, Xiao-Qin
    Yeung, Daniel S.
    PROCEEDINGS OF 2008 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-7, 2008, : 794 - +
  • [45] Bounds on the generalization ability of Bayesian inference and Gibbs algorithms
    Teytaud, O
    Paugam-Moisy, H
    ARTIFICIAL NEURAL NETWORKS-ICANN 2001, PROCEEDINGS, 2001, 2130 : 265 - 270
  • [46] Enhancing the generalization ability of neural networks by using Gram-Schmidt orthogonalization algorithm
    Wan, WS
    Hirasawa, K
    Hu, JL
    Murata, J
    IJCNN'01: INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-4, PROCEEDINGS, 2001, : 1721 - 1726
  • [47] Improving generalization ability of self-generating neural networks through ensemble averaging
    Inoue, H
    Narihisa, H
    KNOWLEDGE DISCOVERY AND DATA MINING, PROCEEDINGS: CURRENT ISSUES AND NEW APPLICATIONS, 2000, 1805 : 177 - 180
  • [48] The Impact of Enriching Electroencephalogram in Spatial Metadata on Interpretability and Generalization Ability of Graph Neural Networks
    L. S. Sidorov
    A. I. Maysuradze
    Pattern Recognition and Image Analysis, 2024, 34 (4) : 1255 - 1263
  • [49] Improving the Generalization Ability of Deep Neural Networks for Cross-Domain Visual Recognition
    Zheng, Jianwei
    Lu, Chao
    Hao, Cong
    Chen, Deming
    Guo, Donghui
    IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS, 2021, 13 (03) : 607 - 620
  • [50] Comparison of Bayesian networks and artificial neural networks for quality detection in a machining process
    Correa, M.
    Bielza, C.
    Pamies-Teixeira, J.
    EXPERT SYSTEMS WITH APPLICATIONS, 2009, 36 (03) : 7270 - 7279
12345