Communication-Constrained Inference and the Role of Shared Randomness

Authors: Jayadev Acharya, Clement Canonne, Himanshu Tyagi

ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We study problems of distribution learning and identity testing in this distributed inference setting and examine the role of shared randomness as a resource. We propose a general purpose simulate-and-infer strategy that uses only private-coin communication protocols and is sample-optimal for distribution learning. This general strategy turns out to be sample-optimal even for distribution testing among private-coin protocols. Interestingly, we propose a public-coin protocol that outperforms simulate-and-infer for distribution testing and is, in fact, sample-optimal.
Researcher Affiliation Academia 1Cornell University 2Stanford University 3Indian Institute of Science Institute of Technology. Correspondence to: Clément Canonne <ccanonne@cs.stanford.edu>.
Pseudocode No The paper describes protocols and schemes in paragraph form, such as 'We describe the base version of the protocol below...' in Theorem 4.1, but it does not include any clearly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not contain any statement about releasing source code for the methodology described, nor does it provide a link to a code repository.
Open Datasets No The paper is theoretical and focuses on statistical inference and communication complexity; it does not describe using any datasets for training or provide access information for publicly available datasets.
Dataset Splits No The paper is theoretical and does not describe experiments with datasets, thus it does not provide details on training, validation, or test dataset splits.
Hardware Specification No The paper is theoretical and does not describe running experiments. Therefore, no hardware specifications are mentioned.
Software Dependencies No The paper is theoretical and focuses on mathematical proofs and algorithm design, not specific software implementations. It does not list any software dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not describe experimental procedures or configurations such as hyperparameters or training schedules.