Code for my multi-armed bandit simulation optimization project. The primary code for usage/generating results is in the remake branch (https://github.com/magikarp01/Gradient-MAB/tree/remake). The paper (soon to be published in Winter Simulation Conference 2022 Proceedings) can be found at https://drive.google.com/file/d/1W0UGG6mx9hYD5YJjecsPp2-2C7_N4avO.
Purpose The purpose of this code is to implement and test the strategies for multi-start gradient descent described in "BANDIT-BASED MULTI-START STRATEGIES FOR GLOBAL CONTINUOUS OPTIMIZATION", a paper by Phillip Guo and Dr. Michael C. Fu.