Large sample analysis of the median heuristic
arXiv:1707.07269
Abstract
In kernel methods, the median heuristic has been widely used as a way of setting the bandwidth of RBF kernels. While its empirical performances make it a safe choice under many circumstances, there is little theoretical understanding of why this is the case. Our aim in this paper is to advance our understanding of the median heuristic by focusing on the setting of kernel two-sample test. We collect new findings that may be of interest for both theoreticians and practitioners. In theory, we provide a convergence analysis that shows the asymptotic normality of the bandwidth chosen by the median heuristic in the setting of kernel two-sample test. Systematic empirical investigations are also conducted in simple settings, comparing the performances based on the bandwidths chosen by the median heuristic and those by the maximization of test power.
27 pages, 6 figures
References in corpus (1)
Cited by in corpus (26)
- Statistical Inference for Generative Models with Maximum Mean Discrepancy
- Adversarial purification with Score-based generative models
- Kernel-based Graph Learning from Smooth Signals: A Functional Viewpoint
- User Training with Error Augmentation for Electromyogram-based Gesture Classification
- Geometric Methods for Sampling, Optimisation, Inference and Adaptive Agents
- Simulator Calibration under Covariate Shift with Kernels
- Optimal quantisation of probability measures using maximum mean discrepancy
- Generalized Kernel Thinning
- Kernelized Wasserstein Natural Gradient
- PEARL: Data Synthesis via Private Embeddings and Adversarial Reconstruction Learning
- Deep IDA: A Deep Learning Method for Integrative Discriminant Analysis of Multi-View Data with Feature Ranking -- An Application to COVID-19 severity
- Positively Weighted Kernel Quadrature via Subsampling
- Improved Autoregressive Modeling with Distribution Smoothing
- Energy distance and kernel mean embeddings for two-sample survival testing
- Projected State-action Balancing Weights for Offline Reinforcement Learning
- Measure Transport with Kernel Stein Discrepancy
- Kernel-Based Testing for Single-Cell Differential Analysis
- A General Framework for Powerful Confounder Adjustment in Omics Association Studies
- Sampling with Mirrored Stein Operators
- Nonparametric Causal Feature Selection for Spatiotemporal Risk Mapping of Malaria Incidence in Madagascar
- Kernel Conditional Density Operators
- Kernel-Guided Training of Implicit Generative Models with Stability Guarantees
- Estimation of Counterfactual Interventions under Uncertainties
- A Critical Reexamination of Intra-List Distance and Dispersion
- Spectral properties of kernel matrices in the flat limit
- Interpretable Stein Goodness-of-fit Tests on Riemannian Manifolds