Algorithms for the Communication of Samples

Authors: Lucas Theis, Noureldin Y Ahmed

ICML 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We run two sets of empirical experiments to compare the reverse channel coding schemes discussed above. We first investigate the effect of hybrid coding on the computational cost of communicating a (truncated) Gaussian sample. We then compare the performance of a wider set of algorithms for the task of approximately simulating a categorical distribution.
Researcher Affiliation Industry 1Google, London, UK 2Google, Dublin, Ireland.
Pseudocode Yes Algorithm 1 PFR; Algorithm 2 Hybrid coding; Algorithm 3 Greedy rejection sampling (RS*; Harsha et al., 2007); Algorithm 4 Ordered random coding (ORC)
Open Source Code No The paper does not provide any links to a code repository or make explicit statements about open-sourcing their code.
Open Datasets No The paper describes generating samples from specified distributions (Gaussian and Dirichlet-distributed categorical distributions) but does not refer to or provide access information for a pre-existing, publicly available dataset in the conventional sense.
Dataset Splits No The paper does not explicitly mention training, validation, or test dataset splits, cross-validation, or any other data partitioning strategy.
Hardware Specification No The paper does not specify any hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., programming languages, libraries, frameworks).
Experiment Setup Yes Appropriate values for wmin and M are provided in Appendix H.