CRVI: Convex Relaxation for Variational Inference
Authors: Ghazal Fazelnia, John Paisley
ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We experiment on 9 datasets from the UCI repository with various sizes and dimensions. These data sets are: Iris, Birth rate and economic growth, Yacht, Pima Indian diabetes, Bike sharing, Parkinson data, Wisconsin breast cancer (WDBC), Online news popularity, Year of release prediction for a million songs. We experimented using 100 different hyper-parameter settings and initial values for each dataset. Table (1) shows some details about these datasets, as well as the average running time for our simulations and the average rank of the optimal solution found by CRVI. |
| Researcher Affiliation | Academia | 1Department of Electrical Engineering & Data Science Institute, Columbia University, New York, USA. |
| Pseudocode | No | The paper describes its methods in prose and mathematical equations but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper mentions implementing CRVI using CVX (a third-party package) but does not provide any statement or link indicating that their own CRVI implementation code is open-source or publicly available. |
| Open Datasets | Yes | We experiment on 9 datasets from the UCI repository with various sizes and dimensions. These data sets are: Iris, Birth rate and economic growth, Yacht, Pima Indian diabetes, Bike sharing, Parkinson data, Wisconsin breast cancer (WDBC), Online news Popularity, Year of release prediction for a million songs. |
| Dataset Splits | No | The paper mentions using several datasets from the UCI repository and synthetic data, but it does not specify any training, validation, or test dataset splits or methodologies. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used for running its experiments. |
| Software Dependencies | Yes | We implemented CRVI code using CVX, which is a package for specifying and solving convex programs (Grant and Boyd, 2014; 2008). |
| Experiment Setup | No | The paper states 'We experimented using 100 different hyper-parameter settings and initial values for each dataset.' but does not provide the specific hyperparameter values or detailed configurations used. |