Intersecting Faces: Non-negative Matrix Factorization With New Guarantees

Authors: Rong Ge, James Zou

ICML 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We explored the performance of Face-Intersect on simulations and discuss settings where it empirically outperformed the state-of-art methods.
Researcher Affiliation Industry Rong Ge RONGGE@MICROSOFT.COM Microsoft Research New England, James Zou JAZO@MICROSOFT.COM Microsoft Research New England
Pseudocode Yes Algorithm 1 Face-Intersect, Algorithm 2 Finding a properly filled facet, Algorithm 3 Finding all proper facets, Algorithm 4 Finding Intersection, Algorithm 5 Finding remaining vertices
Open Source Code No The paper does not provide any specific links or explicit statements indicating that open-source code for the described methodology is available.
Open Datasets No The paper states, "We simulated data according to the generative NMF model described in Section 7," and describes the data generation process, but it does not provide concrete access information (link, DOI, repository, or standard benchmark citation) for a publicly available or open dataset.
Dataset Splits No The paper describes simulating data and experimental settings but does not provide specific details on training, validation, or test dataset splits.
Hardware Specification No The paper does not explicitly describe the specific hardware (e.g., CPU/GPU models, memory) used to run its experiments.
Software Dependencies No The paper mentions algorithms like "Anchor-Words algorithm" and "Projected Gradient" but does not provide specific version numbers for any software dependencies or libraries.
Experiment Setup Yes We tested a range of settings with m between 5 to 100, r between 3 to 10, and n1 and n2 between 100 and 500. We generated the true data as M = AW and added i.i.d. Gaussian noise to each entry of M to generate the observed data M.