Coded Computing for Resilient Distributed Computing: A Learning-Theoretic Framework

Authors: Parsa Moradi, Behrooz Tahmasebi, Mohammad Maddah-Ali

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we extensively evaluate the proposed scheme across various scenarios. Our assessments involve examining multiple deep neural networks as computing functions and exploring the impact of different numbers of stragglers on the scheme s efficiency.
Researcher Affiliation Academia Parsa Moradi University of Minnesota moradi@umn.edu Behrooz Tahmasebi MIT CSAIL bzt@mit.edu Mohammad Ali Maddah-Ali University of Minnesota maddah@umn.edu
Pseudocode No No explicit pseudocode or algorithm block found.
Open Source Code No Question: Does the paper provide open access to the data and code, with sufficient instructions to faithfully reproduce the main experimental results, as described in supplemental material? Answer: [No] Justification: We provided references to all the open datasets that we used in the paper. Regarding the code, we are happy to share it later if required.
Open Datasets Yes Shallow model: We choose Le Net5 [36] architecture as a known shallow network with approximately 6 104 parameters, trained on the MNIST [37]. Deep model with low-dimensional output: ...CIFAR-10 [38] dataset. ...Deep model with high-dimensional output: ...Image Net-1K dataset [41].
Dataset Splits Yes The sole hyper-parameters involved are the two smoothing parameters (λenc, λdec) which are determined using cross-validation and greed search over different values of the smoothing parameters.
Hardware Specification No The experiments are run using Py Torch [35] in a single GPU machine.
Software Dependencies No The experiments are run using Py Torch [35] in a single GPU machine.
Experiment Setup Yes The sole hyper-parameters involved are the two smoothing parameters (λenc, λdec) which are determined using cross-validation and greed search over different values of the smoothing parameters.