Distributed Nonparametric Regression under Communication Constraints

Authors: Yuancheng Zhu, John Lafferty

ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical This paper studies the problem of nonparametric estimation of a smooth function with data distributed across multiple machines. We assume an independent sample from a white noise model is collected at each machine, and an estimator of the underlying true function needs to be constructed at a central machine. We place limits on the number of bits that each machine can use to transmit information to the central machine. Our results give both asymptotic lower bounds and matching upper bounds on the statistical risk under various settings.
Researcher Affiliation Academia 1Department of Statistics, Wharton School, University of Pennsylvania 2Department of Statistics and Data Science, Yale University.
Pseudocode Yes First we state a lemma describing and analyzing a simple scalar quantization method. ... We are now ready to describe the algorithm of estimating θ. α: order of the Sobolev space. c: radius of the Sobolev space. ... At the jth machine (for j = 1, . . . , m), let Ij = n (ms + j)/k : s = 0, . . . ,eb 1 o .
Open Source Code No The paper does not include any explicit statement about releasing source code or a link to a code repository.
Open Datasets No The paper is theoretical and does not involve empirical training on datasets. It studies models such as the 'white noise model' and 'Gaussian sequence model' as theoretical constructs.
Dataset Splits No The paper is theoretical and does not describe any dataset splits for validation. No empirical experiments are conducted.
Hardware Specification No The paper is theoretical and does not describe any specific hardware used for experiments.
Software Dependencies No The paper is theoretical and does not specify any software dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not describe any experimental setup details, hyperparameters, or system-level training settings.