Continuous-time Analysis of Anchor Acceleration

Authors: Jaewook Suh, Jisun Park, Ernest Ryu

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we present an adaptive method inspired by the continuous-time analyses and establish its effectiveness through theoretical analyses and experiments. We now show an experiment with the method of Theorem 7.2 applied to a decentralized compressed sensing problem Shi et al. [63]. We show the results in Figure 2.
Researcher Affiliation Academia Jaewook J. Suh Seoul National University jacksuhkr@snu.ac.kr Jisun Park Seoul National University colleenp0515@snu.ac.kr Ernest K. Ryu Seoul National University ernestryu@snu.ac.kr
Pseudocode No The paper describes algorithms and methods using mathematical notation and text, but it does not contain a clearly labeled pseudocode block or algorithm section.
Open Source Code No The paper does not include an unambiguous statement about releasing code for the methodology or provide a link to a source-code repository.
Open Datasets Yes We solve a compressed sensing problem of Shi et al. [63].
Dataset Splits No The paper uses a compressed sensing problem but does not specify dataset splits (e.g., percentages, sample counts, or explicit predefined splits) for training, validation, or testing.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper mentions applying PG-EXTRA but does not list specific software dependencies with version numbers (e.g., 'Python 3.8, PyTorch 1.9').
Experiment Setup Yes We choose the dimension of signal d = 100, the number of agents n = 20, the number of measurement for each agent mi = 4, ℓ1-regularization parameter ρ = 0.01, and algorithm parameter α = 0.01.