Convergence guarantees for kernel-based quadrature rules in misspecified settings

Authors: Motonobu Kanagawa, Bharath K. Sriperumbudur, Kenji Fukumizu

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental For the algorithm by Bach [2], we conducted simulation experiments to support this observation, by using code available from http://www.di.ens.fr/~fbach/quadrature.html. The setting is what we have described with d = 1, and weights are obtained without regularization as in [2]. The result is shown in Figure 1, where r (= α) denotes the assumed smoothness, and s (= αθ) is the (unknown) smoothness of an integrand. The straight lines are (asymptotic) upper-bounds in Theorem 1 (slope s and intercept fitted for n 24), and the corresponding solid lines are numerical results (both in log-log scales). Averages over 100 runs are shown. The result indeed shows the adaptability of the quadrature rule by Bach for the less smooth functions (i.e. s = 1, 2, 3).
Researcher Affiliation Academia The Institute of Statistical Mathematics, Tokyo 190-8562, Japan Department of Statistics, Pennsylvania State University, University Park, PA 16802, USA
Pseudocode No The paper describes algorithms (e.g., QMC, Bach's algorithm) but does not present any of them in a structured pseudocode block or algorithm box.
Open Source Code No The paper mentions using "code available from http://www.di.ens.fr/~fbach/quadrature.html" for Bach's algorithm, which is a third-party code. It does not state that the authors are releasing their own source code for the methodology described in this paper.
Open Datasets No The paper describes generating data for simulations (e.g., "d = 1", "uniform distribution") rather than using a named public dataset or providing access information for a custom dataset. There is no mention of dataset availability.
Dataset Splits No The paper describes simulation experiments but does not provide specific training, validation, or test dataset splits. The data is generated based on the theoretical setting rather than being split from a pre-existing dataset.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions using "code available from http://www.di.ens.fr/~fbach/quadrature.html" for Bach's algorithm, but it does not specify any version numbers for this or any other software components, libraries, or programming languages.
Experiment Setup Yes The setting is what we have described with d = 1, and weights are obtained without regularization as in [2]... Averages over 100 runs are shown.