This repository contains the implementation of the R-PSI algorithm from our paper "Robust Pareto Set Identification with Contaminated Bandit Feedback" and Algorithm 1 from "Pareto Front Identification from Stochastic Bandit Feedback".
The algorithms are compared on 3 different setups. The results of each setup can be recreated by changing the variable "setting" inside "main.py".
If you use this library in an academic work, please cite our work "Robust Pareto Set Identification with Contaminated Bandit Feedback", Kerem Bozgan, Cem Tekin:
@misc{bozgan,
doi = {10.48550/ARXIV.2206.02666},
url = {https://arxiv.org/abs/2206.02666},
author = {Bozgan, Kerem and Tekin, Cem},
keywords = {Machine Learning (cs.LG), Applications (stat.AP), Machine Learning (stat.ML), FOS: Computer and information sciences, FOS: Computer and information sciences, I.2.6},
title = {Robust Pareto Set Identification with Contaminated Bandit Feedback},
publisher = {arXiv},
year = {2022},
copyright = {arXiv.org perpetual, non-exclusive license}
}