Gradient Coding: Avoiding Stragglers in Distributed Learning
Authors: Rashish Tandon, Qi Lei, Alexandros G. Dimakis, Nikos Karampatziakis
ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We implement our schemes in python (using MPI) to run on Amazon EC2, and show how we compare against baseline approaches in running time and generalization error. |
| Researcher Affiliation | Collaboration | 1Department of Computer Science, University of Texas at Austin, Austin, TX, USA 2Institute for Computational Engineering and Sciences, University of Texas at Austin, Austin, TX, USA 3Department of Electrical and Computer Engineering, University of Texas at Austin, Austin, TX, USA 4Microsoft, Seattle, WA, USA. |
| Pseudocode | Yes | Algorithm 1 Algorithm to compute A |
| Open Source Code | No | The paper states 'We implemented all methods in python using MPI4py (Dalcin et al., 2011), an open source MPI implementation.', which refers to a third-party library used. However, it does not provide any link or explicit statement that the code for their own proposed methodology is open-source or publicly available. |
| Open Datasets | Yes | Next, we trained a logistic regression model on the Amazon Employee Access dataset from Kaggle 1. ... 1https://www.kaggle.com/c/amazon-employee-access-challenge |
| Dataset Splits | No | The paper mentions 'training samples' and evaluating on 'a true holdout' dataset (implied test set), but it does not provide specific details on training/validation/test dataset splits, such as percentages, absolute counts for each split, or methods like cross-validation. |
| Hardware Specification | Yes | Our experiments were performed using two different worker instance types on Amazon EC2: m1.small and t2.micro these are very small, very low-cost EC2 instances. ... For that reason, and to mitigate this additional overhead to some degree, we used a larger master instance of c3.8xlarge in our experiments. |
| Software Dependencies | No | The paper states 'We implemented all methods in python using MPI4py (Dalcin et al., 2011), an open source MPI implementation.', which mentions software components but does not provide specific version numbers for Python, MPI4py, or any other dependencies. |
| Experiment Setup | No | The paper mentions using 'Nesterov's Accelerated Gradient (NAG) with a constant learning rate' for some methods, but it does not provide specific hyperparameter values such as the exact learning rate, batch size, number of epochs, or other detailed training configurations. |