Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Distributed Mean Estimation with Limited Communication
Authors: Ananda Theertha Suresh, Felix X. Yu, Sanjiv Kumar, H. Brendan McMahan
ICML 2017 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We ο¬nally demonstrate the practicality of our algorithms by applying them to distributed Lloyd s algorithm for kmeans and power iteration for PCA. and We demonstrate two applications in the rest of this section. The experiments are performed on the MNIST (d = 1024) and CIFAR (d = 512) datasets. |
| Researcher Affiliation | Industry | 1Google Research, New York, NY, USA 2Google Research, Seattle, WA, USA. |
| Pseudocode | No | Information insufficient. The paper describes algorithms in prose and mathematical notation but does not include any blocks labeled "Pseudocode" or "Algorithm". |
| Open Source Code | No | Information insufficient. The paper does not mention providing open-source code for the described methodology. |
| Open Datasets | Yes | The experiments are performed on the MNIST (d = 1024) and CIFAR (d = 512) datasets. |
| Dataset Splits | No | Information insufficient. The paper mentions the datasets used but does not provide specific details on how they were split into training, validation, or test sets. |
| Hardware Specification | No | Information insufficient. The paper does not provide specific hardware details (e.g., CPU, GPU models, memory) used for running its experiments. |
| Software Dependencies | No | Information insufficient. The paper does not provide specific software dependencies or their version numbers. |
| Experiment Setup | Yes | Here we test two settings: 16 quantization levels and 32 quantization levels. We set both the number of centers and number of clients to 10. The dataset is distributed over 100 clients. |