Robust Budget Pacing with a Single Sample

Authors: Santiago R. Balseiro, Rachitesh Kumar, Vahab Mirrokni, Balasubramanian Sivan, Di Wang

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our main contribution is to show that Dual Follow-The-Regularized-Leader (FTRL) is robust and achieves the near optimal O( T)-regret even when it has access to just one sample from each distribution Pt, dramatically improving over the prior T log T samples from each distribution (Jiang et al., 2020) to achieve O( T)-regret, while still being robust to noise in the sampling distributions.
Researcher Affiliation Collaboration 1DRO, Columbia Business School, New York, NY, USA 2Google Research, New York, NY, USA 3IEOR, Columbia University, New York, NY, USA. Work done as a Student Researcher at Google Research.
Pseudocode Yes Algorithm 1 Learning the Dual and Earning with It
Open Source Code No The paper does not provide any specific link or explicit statement about the availability of source code for the methodology described.
Open Datasets No The paper is theoretical and does not involve empirical evaluation on a public dataset.
Dataset Splits No The paper is theoretical and does not describe experiments with dataset splits.
Hardware Specification No The paper is theoretical and does not describe experiments or specify any hardware used.
Software Dependencies No The paper is theoretical and does not describe experiments that would require specific software dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not describe an experimental setup with specific hyperparameters or training configurations.