Skip to content

Latest commit

 

History

History
13 lines (10 loc) · 845 Bytes

README.md

File metadata and controls

13 lines (10 loc) · 845 Bytes

Perturbed Top-K Optimization

Here we present the Perturbed Top-K Optimization objective with extensions to the synthetic data experiments in the IMLH 2023 paper Learning where to intervene with a differentiable top-K operator: Towards data-driven strategies to prevent fatal opioid overdoses.

Requirements:

Numpy, Tensorflow 2.0+, perturbations.py from this repository

Code overview

  • Perturbed Top-K Demonstration.ipynb A notebook demonstrating the creation of synthetic data and model training
  • create_data.py Generates and saves the synthetic data
  • create_model.py Creates a tensorflow model that can utilize the Perturbed-Top-K module, or without
  • data_utils.py Miscellaneous utilities for generating and visualizing the synthetic data