Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Pre-training of Graph Augmented Transformers for Medication Recommendation
Authors: Junyuan Shang, Tengfei Ma, Cao Xiao, Jimeng Sun
IJCAI 2019 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We used EHR data from MIMIC-III [Johnson et al., 2016] and conducted all our experiments on a cohort where patients have more than one visit. ... Table. 3 compares the performance on the medication recommendation task. |
| Researcher Affiliation | Collaboration | Junyuan Shang1,3 , Tengfei Ma2 , Cao Xiao1 and Jimeng Sun3 1Analytics Center of Excellence, IQVIA, Cambridge, MA, USA 2IBM Research AI, Yorktown Heights, NY, USA 3Georgia Institute of Technology, Atlanta, GA, USA |
| Pseudocode | No | The paper describes the method using mathematical equations and descriptive text, but it does not include a clearly labeled pseudocode block or algorithm. |
| Open Source Code | Yes | 1https://github.com/jshang123/G-Bert |
| Open Datasets | Yes | We used EHR data from MIMIC-III [Johnson et al., 2016] |
| Dataset Splits | Yes | We randomly divide the dataset into training, validation and testing set in a 0.6 : 0.2 : 0.2 ratio. |
| Hardware Specification | Yes | All methods are implemented in Py Torch [Paszke et al., 2017] and trained on an Ubuntu 16.04 with 8GB memory and Nvidia 1080 GPU. |
| Software Dependencies | No | The paper mentions 'Py Torch' and 'Ubuntu 16.04' but does not specify version numbers for PyTorch or any other key software libraries. |
| Experiment Setup | Yes | For G-BERT, the hyperparameters are adjusted on evaluation set: (1) GAT part: input embedding dimension as 75, number of attention heads as 4; (2) BERT part: hidden dimension as 300, dimension of position-wise feed-forward networks as 300, 2 hidden layers with 4 attention heads for each layer. ... Training is done through Adam [Kingma and Ba, 2014] at learning rate 5e-4. |