Spectral Methods for Indian Buffet Process Inference

Authors: Hsiao-Yu Tung, Alexander J Smola

NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 6 Experiments We evaluate the algorithm on a number of problems suitable for the two models of (2) and (3). The problems are largely identical to those put forward in [18] in order to keep our results comparable with a more traditional inference approach. We demonstrate that our algorithm is faster, simpler, and achieves comparable or superior accuracy. ... Figure 1 shows that our algorithm is faster and comparatively accurate.
Researcher Affiliation Collaboration Hsiao-Yu Fish Tung Machine Learning Department Carnegie Mellon University Pittsburgh, PA 15213 Alexander J. Smola Machine Learning Department Carnegie Mellon University and Google Pittsburgh, PA 15213
Pseudocode Yes Algorithm 1 Excess Correlation Analysis for Linear-Gaussian model with IBP prior
Open Source Code No The paper does not provide an explicit statement or link for open-source code availability for the described methodology.
Open Datasets Yes For a more realistic analysis we used a microarray dataset. The data consisted of 587 mouse liver samples detecting 8565 gene probes, available as dataset GSE2187 as part of NCBI s Gene Expression Omnibus www.ncbi.nlm.nih.gov/geo.
Dataset Splits No The paper mentions sample sizes and training on N=500 samples, but does not explicitly provide training/validation/test dataset splits or cross-validation details.
Hardware Specification No The paper does not provide any specific hardware specifications used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes Using an additive noise variance of σ2 = 0.5 we are able to recover the original signal quite accurately... We used 10 initial iterations 50 random seeds and 30 final iterations 50 in the Robust Power Tensor Method.