AISTATS 2020poster8 citations
Derivative-Free & Order-Robust Optimisation
Haitham Ammar, Victor Gabillon, Rasul Tutunov, Michal Valko
Abstract
In this paper, we formalise order-robust optimisation as an instance of online learning minimising simple regret, and propose Vroom, a zero’th order optimisation algorithm capable of achieving vanishing regret in non-stationary environments, while recovering favorable rates under stochastic reward-generating processes. Our results are the first to target simple regret definitions in adversarial scenarios unveiling a challenge that has been rarely considered in prior work.
BibTeX
@InProceedings{pmlr-v108-ammar20a,
title = {Derivative-Free & Order-Robust Optimisation},
author = {Ammar, Haitham and Gabillon, Victor and Tutunov, Rasul and Valko, Michal},
booktitle = {Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics},
pages = {2293--2303},
year = {2020},
editor = {Chiappa, Silvia and Calandra, Roberto},
volume = {108},
series = {Proceedings of Machine Learning Research},
month = {26--28 Aug},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v108/ammar20a/ammar20a.pdf},
url = {https://proceedings.mlr.press/v108/ammar20a.html},
abstract = { In this paper, we formalise order-robust optimisation as an instance of online learning minimising simple regret, and propose Vroom, a zero’th order optimisation algorithm capable of achieving vanishing regret in non-stationary environments, while recovering favorable rates under stochastic reward-generating processes. Our results are the first to target simple regret definitions in adversarial scenarios unveiling a challenge that has been rarely considered in prior work.}
}