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].
DeEPCA: Decentralized Exact PCA with Linear Convergence Rate
Authors: Haishan Ye, Tong Zhang
JMLR 2021 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experiments validate the advantages of De EPCA empirically. ... In this section, we will provide empirical studies. ... We conduct experiments on the datasets w8a and a9a which can be downloaded in libsvm datasets. ... We report experiment results in Figure 1 and Figure 2. |
| Researcher Affiliation | Academia | Haishan Ye EMAIL Center for Intelligent Decision-Making and Machine Learning School of Management Xi an Jiaotong University Xi an, China. Tong Zhang EMAIL Computer Science & Mathematics Hong Kong University of Science and Technology Clear Water Bay, Kowloon, Hong Kong. |
| Pseudocode | Yes | Algorithm 1 Decentralized Exact PCA (De EPCA) ... Algorithm 2 Sign Adjust ... Algorithm 3 Fast Mix |
| Open Source Code | No | The paper does not contain any explicit statement about the release of source code for the methodology described, nor does it provide a link to a code repository. |
| Open Datasets | Yes | We conduct experiments on the datasets w8a and a9a which can be downloaded in libsvm datasets. |
| Dataset Splits | No | The paper describes the dimensions of the datasets (n=800, d=300 for w8a; n=600, d=123 for a9a) and how data is distributed among agents. However, it does not provide specific training, validation, or test dataset splits (e.g., percentages or counts) typically used for model evaluation. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used to run the experiments, such as CPU or GPU models, memory, or types of computing resources. |
| Software Dependencies | No | The paper does not specify any software dependencies or their version numbers (e.g., programming languages, libraries, frameworks, or solvers) used for implementing the algorithms or conducting experiments. |
| Experiment Setup | Yes | In our experiments, we consider random networks where each pair of agents has a connection with a probability of p = 0.5. We set L = I M λmax(M) where M is the Laplacian matrix associated with a weighted graph. We set m = 50 , that is, there exists 50 agents in this network. In our experiments, the gossip matrix L satisfies 1 λ2(L) = 0.4563. ... We will study how consensus steps affect the convergence rate of De EPCA empirically. Thus, we set different K s in our experiment. |