Tensor Decomposition via Joint Matrix Schur Decomposition

Authors: Nicolo Colombo, Nikos Vlassis

ICML 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We empirically demonstrate that our algorithm is faster and at least as accurate and robust than state-of-the-art algorithms for this problem. 4. Experiments 4.1. Comparison on synthetic data Figure 1. Decomposition of a symmetric nonorthogonal tensor.
Researcher Affiliation Collaboration Nicol o Colombo NICOLO.COLOMBO@UNI.LU Luxembourg Centre for Systems Biomedicine, University of Luxembourg, Esch-sur Alzette, Luxembourg Nikos Vlassis VLASSIS@ADOBE.COM Adobe Research, San Jose, CA
Pseudocode No The paper describes the Gauss-Newton algorithm mathematically and verbally in Section 3, but it does not include a clearly labeled pseudocode or algorithm block.
Open Source Code No The paper does not provide any explicit statement or link indicating that the source code for the proposed methodology is publicly available.
Open Datasets Yes To test the performance of our algorithm on real-world data we have chosen a label prediction problem from crowdsourcing data. The problem and the dataset are described by Zhang et al. (2014) where an estimator based on orderthree moments is also proposed.
Dataset Splits No The paper discusses generating synthetic data and using a real-world dataset from a cited work, but it does not specify explicit training, validation, or test dataset splits (e.g., percentages, counts, or predefined split references).
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions using 'Matlab codes' for comparison methods but does not provide specific version numbers for Matlab or any other software dependencies used in the experiments.
Experiment Setup No The paper describes the algorithmic steps and initialization strategy for the Gauss-Newton algorithm, but it does not provide specific numerical values for hyperparameters (e.g., learning rate, batch size, number of iterations/epochs) or detailed convergence criteria for the experimental setup.