Gaussian Process Surrogate Models for Neural Networks

被引:0
|
作者
Li, Michael Y. [1 ]
Grant, Erin [2 ]
Griffiths, Thomas L. [3 ]
机构
[1] Stanford Univ, Dept Comp Sci, Stanford, CA 94305 USA
[2] UCL, Gatsby Computat Neurosci Unit, London, England
[3] Dept Psychol & Comp Sci, Princeton, NJ USA
来源
关键词
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
Not being able to understand and predict the behavior of deep learning systems makes it hard to decide what architecture and algorithm to use for a given problem. In science and engineering, modeling is a methodology used to understand complex systems whose internal processes are opaque. Modeling replaces a complex system with a simpler, more interpretable surrogate. Drawing inspiration from this, we construct a class of surrogate models for neural networks using Gaussian processes. Rather than deriving kernels for infinite neural networks, we learn kernels empirically from the naturalistic behavior of finite neural networks. We demonstrate our approach captures existing phenomena related to the spectral bias of neural networks, and then show that our surrogate models can be used to solve practical problems such as identifying which points most influence the behavior of specific neural networks and predicting which architectures and algorithms will generalize well for specific datasets.
引用
收藏
页码:1241 / 1252
页数:12
相关论文
共 50 条
  • [21] Neural Dynamics Discovery via Gaussian Process Recurrent Neural Networks
    She, Qi
    Wu, Anqi
    35TH UNCERTAINTY IN ARTIFICIAL INTELLIGENCE CONFERENCE (UAI 2019), 2020, 115 : 454 - 464
  • [22] Adaptive Gaussian process regression for efficient building of surrogate models in inverse problems
    Semler, Phillip
    Weiser, Martin
    INVERSE PROBLEMS, 2023, 39 (12)
  • [23] Bayesian Logistic Gaussian Process Models for Dynamic Networks
    Durante, Daniele
    Dunson, David B.
    ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 33, 2014, 33 : 194 - 201
  • [24] Optimization of magnetically driven directional solidification of silicon using artificial neural networks and Gaussian process models
    Dropka, Natasha
    Holena, Martin
    JOURNAL OF CRYSTAL GROWTH, 2017, 471 : 53 - 61
  • [25] Gravitational-wave surrogate models powered by artificial neural networks
    Khan, Sebastian
    Green, Rhys
    PHYSICAL REVIEW D, 2021, 103 (06)
  • [26] Surrogate Models of Hydrogen Oxidation Kinetics based on Deep Neural Networks
    Akeweje, E.
    Vanovskiy, V.
    Vishnyakov, A.
    THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING, 2023, 57 (02) : 196 - 204
  • [27] Physics-informed neural networks as surrogate models of hydrodynamic simulators
    Donnelly, James
    Daneshkhah, Alireza
    Abolfathi, Soroush
    SCIENCE OF THE TOTAL ENVIRONMENT, 2024, 912
  • [28] Surrogate Models of Hydrogen Oxidation Kinetics based on Deep Neural Networks
    E. Akeweje
    V. Vanovskiy
    A. Vishnyakov
    Theoretical Foundations of Chemical Engineering, 2023, 57 : 196 - 204
  • [29] NEURAL NETWORKS SURROGATE MODELS FOR SIMULATING PAYMENT RISK IN PAVEMENT CONSTRUCTION
    Manik, Anshu
    Gopalakrishnan, Kasthurirangan
    Singh, Abhishek
    Yan, Shengquan
    JOURNAL OF CIVIL ENGINEERING AND MANAGEMENT, 2008, 14 (04) : 235 - 240
  • [30] Gaussian mixture models for training Bayesian convolutional neural networks
    Mostafa, Bakhouya
    Hassan, Ramchoun
    Mohammed, Hadda
    Tawfik, Masrour
    EVOLUTIONARY INTELLIGENCE, 2024, 17 (04) : 2515 - 2536