Performance analysis of greedy algorithms for minimising a maximum mean discrepancy

DOI10.1007/s11222-022-10184-1zbMath1499.62029arXiv2101.07564OpenAlexW3124611209MaRDI QIDQ2104022

Luc Pronzato

Publication date: 9 December 2022

Published in: Statistics and Computing (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2101.07564

zbMATH Keywords

greedy algorithmcomputer experimentsspace-filling designmaximum mean discrepancyquantisationkernel herdingsequential Bayesian quadrature

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08)

Related Items (1)

BLUE against OLSE in the location model: energy minimization and asymptotic considerations

Uses Software

• energy

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Equivalence of distance-based and RKHS-based statistics in hypothesis testing
• An asymptotically optimal gradient algorithm for quadratic optimization with low computational cost
• Energy statistics: a class of statistics based on distances
• On energy, discrepancy and group invariant measures on measurable subsets of Euclidean space
• Design of computer experiments: space filling and beyond
• Conditional gradient algorithms with open loop step size rules
• Design of experiments in nonlinear models. Asymptotic normality, optimality criteria and small-sample properties
• Foundations of quantization for probability distributions
• Support points
• Minimum-energy measures for singular kernels
• Measuring sample quality with diffusions
• On the positivity and magnitudes of Bayesian quadrature weights
• Probabilistic integration: a role in statistical computation?
• Coordinate descent algorithms
• On Khachiyan's algorithm for the computation of minimum-volume enclosing ellipsoids
• Sequences converging to D-optimal designs of experiments
• Hilbert space embeddings and metrics on probability measures
• Approximation Theorems of Mathematical Statistics
• Convergence Rates for Conditional Gradient Sequences Generated by Implicit Step Length Rules
• Finding the nearest point in A polytope
• A generalized discrepancy and quadrature error bound
• Bayesian Quadrature, Energy Minimization, and Space-Filling Design
• Maximum projection designs for computer experiments
• Linear convergence of a modified Frank–Wolfe algorithm for computing minimum-volume enclosing ellipsoids
• The Sequential Generation of $D$-Optimum Experimental Designs
• Control Functionals for Monte Carlo Integration

This page was built for publication: Performance analysis of greedy algorithms for minimising a maximum mean discrepancy