A prediction rigidity formalism for low-cost uncertainties in trained neural networks

被引:0
|
作者
Bigi, Filippo [1 ]
Chong, Sanggyu [1 ]
Ceriotti, Michele [1 ]
Grasselli, Federico [1 ,2 ]
机构
[1] Ecole Polytech Fed Lausanne, Inst Materiaux, Lab Computat Sci & Modeling, CH-1015 Lausanne, Switzerland
[2] Univ Modena & Reggio Emilia, Dept Phys Informat & Math, via Giuseppe Campi 213-a, I-41125 Modena, Italy
来源
MACHINE LEARNING-SCIENCE AND TECHNOLOGY | 2024年 / 5卷 / 04期
基金
瑞士国家科学基金会; 欧洲研究理事会; 欧盟地平线“2020”;
关键词
predictions; rigidity; low-cost uncertainties; neural network; regression; pre-trained; uncertainty quantification;
D O I
10.1088/2632-2153/ad805f
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Quantifying the uncertainty of regression models is essential to ensure their reliability, particularly since their application often extends beyond their training domain. Based on the solution of a constrained optimization problem, this work proposes 'prediction rigidities' as a formalism to obtain uncertainties of arbitrary pre-trained regressors. A clear connection between the suggested framework and Bayesian inference is established, and a last-layer approximation is developed and rigorously justified to enable the application of the method to neural networks. This extension affords cheap uncertainties without any modification to the neural network itself or its training procedure. The effectiveness of this approach is shown for a wide range of regression tasks, ranging from simple toy models to applications in chemistry and meteorology.
引用
收藏
页数:20
相关论文
共 50 条
  • [31] A Low-Cost Computational Spectrometer Based on a Trained Sparse Base Matrix
    Gao, Yanbo
    Pan, Hejia
    Sheng, Yajuan
    Wen, Rui
    Zheng, Yuanhao
    Yang, Lin
    SENSORS, 2025, 25 (03)
  • [32] LOW-COST NETWORKS AID DATA ACQUISITION
    MACDONALD, N
    CONTROL AND INSTRUMENTATION, 1985, 17 (11): : 87 - &
  • [33] An analysis of European low-cost airlines and their networks
    Dobruszkes, Frederic
    JOURNAL OF TRANSPORT GEOGRAPHY, 2006, 14 (04) : 249 - 264
  • [34] Toward Low-Cost and Stable Blockchain Networks
    Fang, Minghong
    Liu, Jia
    ICC 2020 - 2020 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC), 2020,
  • [35] Multimedia QoS in low-cost home networks
    Scholten, H
    Jansen, P
    Hanssen, F
    Hartel, P
    Hattink, T
    Sundramoorthy, V
    LCN 2002: 27TH ANNUAL IEEE CONFERENCE ON LOCAL COMPUTER NETWORKS, PROCEEDINGS, 2002, : 327 - 328
  • [36] ZigBee paves the way for low-cost networks
    Cravotta, N
    EDN, 2004, 49 (21) : 60 - 61
  • [37] Treating uncertainties in reservoir-performance prediction with neural networks
    Lechner, J.P.
    Zangl, G.
    JPT, Journal of Petroleum Technology, 2006, 58 (06): : 69 - 71
  • [38] The impact of low-cost molecular geometry optimization in property prediction via graph neural network
    Pinheiro, Gabriel A.
    Calderan, Felipe V.
    Da Silva, Juarez L. F.
    Quiles, Marcos G.
    2022 21ST IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS, ICMLA, 2022, : 603 - 608
  • [39] Low-cost delamination monitoring of CFRP beams using electrical resistance changes with neural networks
    Todoroki, Akira
    Ueda, Masahito
    SMART MATERIALS AND STRUCTURES, 2006, 15 (04) : N75 - N84
  • [40] Experiments with Neural Networks in the Identification and Control of a Magnetic Levitation System Using a Low-Cost Platform
    Silva, Bruno E.
    Barbosa, Ramiro S.
    APPLIED SCIENCES-BASEL, 2021, 11 (06):