Variational Inference for Sparse and Undirected Models

Authors: John Ingraham, Debora Marks

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

Reproducibility Variable Result LLM Response
Research Type Experimental We find that, together, these methods for variational inference substantially improve learning of sparse undirected graphical models in simulated and real problems from physics and biology.
Researcher Affiliation Academia John Ingraham 1 Debora Marks 1 1Harvard Medical School, Boston, Massachusetts.
Pseudocode Yes Algorithm 1 Computing ELBO for Fadeout
Open Source Code No The paper does not provide a specific link or explicit statement about the availability of its own source code for the described methodology.
Open Datasets No We generated synthetic couplings for two kinds of Ising systems... We sampled synthetic data for each system with the Swendsen-Wang algorithm (Appendix) (Swendsen & Wang, 1987). and We applied the hierarchical Bayesian model from the protein simulation to model across-species amino acid covariation in the SH3 domain family (Figure 6). The paper does not provide concrete access information for the datasets used.
Dataset Splits No No specific details on train/validation/test splits, such as percentages or sample counts for each partition, were provided for full reproducibility, although a test set is mentioned.
Hardware Specification No Portions of this work were conducted on the Orchestra HPC Cluster at Harvard Medical School.
Software Dependencies No The paper does not provide specific software dependency details with version numbers (e.g., Python 3.8, PyTorch 1.9).
Experiment Setup No The paper does not provide specific details on hyperparameters, training configurations, or system-level settings for the experiments.