Convergence Guarantees for Adaptive Bayesian Quadrature Methods

被引:0
|
作者
Kanagawa, Motonobu [1 ,2 ,3 ]
Hennig, Philipp [2 ,3 ]
机构
[1] EURECOM, Sophia Antipolis, France
[2] Univ Tubingen, Tubingen, Germany
[3] Max Planck Inst Intelligent Syst, Tubingen, Germany
基金
欧洲研究理事会;
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
Adaptive Bayesian quadrature (ABQ) is a powerful approach to numerical integration that empirically compares favorably with Monte Carlo integration on problems of medium dimensionality (where non-adaptive quadrature is not competitive). Its key ingredient is an acquisition function that changes as a function of previously collected values of the integrand. While this adaptivity appears to be empirically powerful, it complicates analysis. Consequently, there are no theoretical guarantees so far for this class of methods. In this work, for a broad class of adaptive Bayesian quadrature methods, we prove consistency, deriving non-tight but informative convergence rates. To do so we introduce a new concept we call weak adaptivity. Our results identify a large and flexible class of adaptive Bayesian quadrature rules as consistent, within which practitioners can develop empirically efficient methods.
引用
收藏
页数:12
相关论文
共 50 条
  • [1] Stochastic Convergence Rates and Applications of Adaptive Quadrature in Bayesian Inference
    Bilodeau, Blair
    Stringer, Alex
    Tang, Yanbo
    JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2024, 119 (545) : 690 - 700
  • [2] Convergence guarantees for generalized adaptive stochastic search methods for continuous global optimization
    Regis, Rommel G.
    EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2010, 207 (03) : 1187 - 1202
  • [3] Convergence guarantees for kernel-based quadrature rules in misspecified settings
    Kanagawa, Motonobu
    Sriperumbudur, Bharath K.
    Fukumizu, Kenji
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 29 (NIPS 2016), 2016, 29
  • [4] Frank-Wolfe Bayesian Quadrature: Probabilistic Integration with Theoretical Guarantees
    Briol, Francois-Xavier
    Oates, Chris J.
    Girolami, Mark
    Osborne, Michael A.
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 28 (NIPS 2015), 2015, 28
  • [5] On the convergence of Bayesian Adaptive Filtering
    Sadiki, T
    Slock, DTM
    ISSPA 2005: The 8th International Symposium on Signal Processing and its Applications, Vols 1 and 2, Proceedings, 2005, : 779 - 782
  • [6] ADAPTIVE QUADRATURE - CONVERGENCE OF PARALLEL AND SEQUENTIAL ALGORITHMS
    RICE, JR
    BULLETIN OF THE AMERICAN MATHEMATICAL SOCIETY, 1974, 80 (06) : 1250 - 1254
  • [7] Disentangling Complexity in Bayesian Automatic Adaptive Quadrature
    Adam, Gheorghe
    Adam, Sanda
    MATHEMATICAL MODELING AND COMPUTATIONAL PHYSICS 2017 (MMCP 2017), 2018, 173
  • [8] The Boundary Layer Problem in Bayesian Adaptive Quadrature
    Adam, Gh.
    Adam, S.
    PHYSICS OF PARTICLES AND NUCLEI LETTERS, 2008, 5 (03) : 269 - 273
  • [9] Control Decisions in the Bayesian Automatic Adaptive Quadrature
    Adam, Gh
    Adam, S.
    PROCEEDINGS 2015 8TH ROMANIA TIER 2 FEDERATION GRID, CLOUD & HIGH PERFORMANCE COMPUTING IN SCIENCE (RO-LCG), 2015,
  • [10] Handling accuracy in Bayesian automatic adaptive quadrature
    Adam, Gh
    Adam, S.
    IWTCP-2 AND NCTP-39, 2015, 627