A Birth-Death Process for Feature Allocation

Authors: Konstantina Palla, David Knowles, Zoubin Ghahramani

ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 6. Experiments We experimentally evaluate the BEP model on real-world genomics and social network data. To evaluate the model fit, we compared the BEP model to independent models at each time point.
Researcher Affiliation Collaboration 1University of Oxford, Oxford, UK 2Stanford University, California, USA 3University of Cambridge, Cambridge, UK 4Uber AI Labs, SF, California, USA.
Pseudocode No The paper describes the model and processes mathematically but does not include structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any explicit statements about making the source code available or provide links to a code repository.
Open Datasets Yes Here we used a subset of the gene expression data from Piechota et al. (2010)..., For this experiment we used Ch IP-seq (chromatin immunoprecipitation sequencing) data downloaded from the ENCODE project (Consortium, 2007), In van de Bunt et al. (1999), 32 university freshman students...
Dataset Splits Yes We created 7 train-test splits holding out 20% of the data... and We ran 7 different held-out tests, holding out a different 20% of the data each time. and holding out 10% of all links across all time points.
Hardware Specification No The paper does not provide any specific details about the hardware used for running the experiments.
Software Dependencies No The paper mentions employing Markov Chain Monte Carlo (MCMC) for posterior inference but does not provide specific software names with version numbers for implementation or dependencies.
Experiment Setup Yes We created 7 train-test splits holding out 20% of the data, and ran 700 MCMC iterations. and a burnin of 500. and We choose a Gaussian prior over A, i.e Afm N(0, 1). and we assume the priors wt(k, l) N µw, σ2 w and s N µs, σ2 s .