Code for the experiments run for our Paper: Customizing ML Predictions for Online Algorithms [ICML 2020]
Use Jupyter to run the .ipynb noteboook. The number of experiments and the size of the data-set can be changed inside the code.
General vs Margin:
The first few cells have the functions that we need to run the experiment. Then the next cells run the experiments for the General PAC learning method, followed by the margin method. The plots are displayed at the end.
PAC Learning Noise:
Run each cell individually and the plots are displayed at the end.