** Code for Learning to Rank for Optimal Treatment Allocation Under Resource Constraints **

Code for Learning to Rank for Optimal Treatment Allocation Under Resource Constraints, published at AIStats 2024.

Instructions:

1. Install required packages in requirements.txt
2. Set relevant hyperparameters in run_models.py. a. method = 1 runs a random forest splitting based on AUTOC, whereas method = 0 runs a baseline model that splits to maximize MSE. b. Other relevant hyperparameters are
3. Run run_models.py a. Models and results will be saved out in relevant directories, with unique identifiers based on relevant parameters. These can be adjusted as needed.