This repository provides an implementation in python of a fair and diverse sampling mechanism based on DPPs (Determinantal Point Processes) along with experiments comparing several exising sampling methods as presented in our ICML'18 paper (available here https://arxiv.org/abs/1802.04023).
- If you want to see a simple demonstration of DPP sampling see "example_sampling.py".
- If you want to run the experiments using the image data-set run "run_image_experiment.py".
- If you want to run the experiments using the adult data-set run "run_adult_experiment.py".
Fair and Diverse DPP-Based Data Summarization
L. Elisa Celis, Vijay Keswani, Damian Straszak, Amit Deshpande, Tarun Kathuria, Nisheeth K. Vishnoi
International Conference on Machine Learning, ICML 2018