Distance Based Modeling of Interactions in Structured Regression
Authors: Ivan Stojkovic, Vladisav Jelisavcic, Veljko Milutinovic, Zoran Obradovic
IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | 3 Experimental Results In order to characterize the improved capabilities of the extended model over the baseline model we have conducted a number of computational experiments described in the following two subsections. 3.1 Synthetic data ... 3.2 Real Applications ... |
| Researcher Affiliation | Academia | Ivan Stojkovic,1,3 Vladisav Jelisavcic,2,3 Veljko Milutinovic,3 and Zoran Obradovic1 1Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, PA, USA 2Mathematical Institute, Serbian Academy of Sciences and Arts, Belgrade, Serbia 3School of Electrical Engineering, University of Belgrade, Belgrade, Serbia |
| Pseudocode | No | The paper does not contain any sections labeled "Pseudocode" or "Algorithm", nor does it present structured algorithm blocks. |
| Open Source Code | No | The paper does not contain any explicit statements about releasing source code for the described methodology or links to code repositories. |
| Open Datasets | Yes | Sepsis Admissions Prediction in California Hospitals (SEPSIS). The monthly admission rate is predicted for sepsis in the California hospitalization dataset [(SID), 2003 2011], which contained admission information for 108 months at 231 hospitals. Data for this experiment was provided by the Health Care and Utilization Project (HCUP) and State Inpatient Databases (SID). ... Precipitation Estimation in Continental US (RAIN). The rain dataset contains precipitation records from meteorological stations across the USA and has been acquired from NOAA s National Climate Data Center (NCDC) [Menne et al., 2009]. ... (data available on NOAA website: http://www.esrl.noaa.gov/psd/). |
| Dataset Splits | Yes | At each node in a graph representing 231 hospitals in California we trained unstructured predictors using data from the initial 75 months. This is repeated 10 times by sampling 50 out of 75 months at random where remaining 25 months were used to learn the parameters of a structured model. The future 30 months were used as a test set in each of ten experiments. ... Predictive models were trained by randomly selecting 250 months from the initial 400 months repeated 10 times, while structured models were learned on the remaining 150 months and performance was assessed on the future 308 months. |
| Hardware Specification | No | The paper does not specify any particular hardware components (e.g., CPU, GPU models, memory, etc.) used for conducting the experiments. |
| Software Dependencies | No | The paper mentions using "neural network models (NN)" and "Gaussian Process Regression (GPR) with Gaussian kernel" but does not provide specific version numbers for these or any other software dependencies. |
| Experiment Setup | No | The paper describes the general setup of training data usage and model application (e.g., "sampled 250 instances... for training the neural network models"), but it does not provide specific hyperparameter values (e.g., learning rate, batch size, number of epochs, optimizer settings) or detailed training configurations for the models used. |