Federated Spectral Clustering via Secure Similarity Reconstruction
Authors: Dong Qiao, Chris Ding, Jicong Fan
NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Numerical results on synthetic and real datasets demonstrate that the proposed method is efficient and accurate in comparison to baselines. and We test our method on both synthetic data and real datasets in comparison to baselines, which verify the effectiveness of our method. |
| Researcher Affiliation | Academia | Dong Qiao1,2 Chris Ding1 Jicong Fan 1,2 1The Chinese University of Hong Kong, Shenzhen, China 2Shenzhen Research Institute of Big Data, Shenzhen, China dongqiao@link.cuhk.edu.cn {chrisding,fanjicong}@cuhk.edu.cn |
| Pseudocode | Yes | Algorithm 1 Proposed Federated Similarity Reconstruction |
| Open Source Code | No | The paper does not provide any specific links to source code or explicitly state that the code for their methodology is being released. |
| Open Datasets | Yes | Taking the COIL20 dataset [Nene et al., 1996] as an example..., We compare our method with the clustering method DSC proposed by [Hernández-Pereira et al., 2021]...Iris [Dua and Graff, 2017], COIL20 [Nene et al., 1996], banknote authentication [Dua and Graff, 2017], and USPS [Hull, 1994]., and Both iris and banknote authentication are from the UCI machine learning library. COIL20 is an image dataset from Columbia Imaging and Vision Laboratory. USPS is a dataset for handwritten text recognition research. |
| Dataset Splits | No | The paper describes data distribution across P clients and mentions train for their model (e.g., next round of training) but does not specify explicit percentages or counts for training, validation, and test dataset splits, nor does it refer to predefined splits from cited sources for reproducibility. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used to run its experiments. |
| Software Dependencies | No | The paper does not provide specific software dependency details (e.g., library or solver names with version numbers) needed to replicate the experiment. |
| Experiment Setup | Yes | In our experiments, we set some hyperparameters including λC, d, r, and k for implementing the proposed Fed SC. Among them, λC as the penalty parameter of the regularization term is set to 1e 2. k is the hyperparameter of the KNN-based operation on the similarity matrix. We set k to max(ceil(log(n)), 1)...Setting of hyparameter r...Setting of hyparameter d... |