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].
Ranking and synchronization from pairwise measurements via SVD
Authors: Alexandre d'Aspremont, Mihai Cucuringu, Hemant Tyagi
JMLR 2021 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We provide a detailed theoretical analysis in terms of robustness against both sampling sparsity and noise perturbations with outliers, using results from matrix perturbation and random matrix theory. Our theoretical findings are complemented by a detailed set of numerical experiments on both synthetic and real data, showcasing the competitiveness of our proposed algorithms with other state-of-the-art methods. |
| Researcher Affiliation | Academia | Alexandre d Aspremont EMAIL CNRS & Ecole Normale Sup erieure, Paris, France; Mihai Cucuringu EMAIL Department of Statistics and Mathematical Institute University of Oxford The Alan Turing Institute, London, UK; Hemant Tyagi EMAIL Inria, Univ. Lille, CNRS UMR 8524 Laboratoire Paul Painlev e, F-59000 |
| Pseudocode | Yes | Algorithm 1 SVD-RS; Algorithm 2 SVD-NRS |
| Open Source Code | No | The paper mentions using the TFOCS software library (Becker et al.) but does not provide specific access to the authors' implementation code for the methodology described in this paper. |
| Open Datasets | Yes | Our second real world example is a set of two networks of animal dominance among captive monk parakeets (Hobson and De Deo, 2015). The input matrix H is skew-symmetric, with Hij denoting the number of net aggressive wins of animal i toward animal j. [...] Our next example covers three North American academic hiring networks, that track the flow of academics between universities (Clauset et al., 2015). |
| Dataset Splits | No | The paper describes generating synthetic data by fixing the number of nodes (n), and varying edge density (p) and noise level (γ). For real datasets, it states that 'Each separate season provided a pairwise comparison matrix' or similar, implying full datasets are used for evaluation, without specifying explicit train/test/validation splits. |
| Hardware Specification | No | The paper does not explicitly describe any hardware used to run its experiments, such as specific GPU or CPU models. |
| Software Dependencies | No | The paper mentions the 'TFOCS software library (Becker et al.)' but does not provide a specific version number. No other software dependencies are listed with version numbers. |
| Experiment Setup | Yes | We fix the number of nodes (n), and vary the edge density (p), and noise level (γ) as follows. We broadly consider two main experimental settings. In Figure 2 (uniform scores) and Figure 3 (Gamma scores) we consider a synthetic model with n = 1000, with sparsity parameters p = 0.05 (column 1) and p = 1 (column 3). [...] In Figure 4 (uniform scores) and Figure 5 (Gamma scores), we consider a synthetic model with n = 3000, and sparsity parameter p {0.01, 0.05, 0.1}, indexing the columns. |