Effective Dimension in Bandit Problems under Censorship

Authors: Gauthier Guinet, Saurabh Amin, Patrick Jaillet

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our main contributions include the introduction of a broad class of censorship models and their analysis in terms of the effective dimension of the problem a natural measure of its underlying statistical complexity and main driver of the regret bound. Our analysis involves a continuous generalization of the Elliptical Potential Inequality, which we believe is of independent interest.
Researcher Affiliation Collaboration Gauthier Guinet AWS AI Labs guinetgg@amazon.com Saurabh Amin MIT amins@mit.edu Patrick Jaillet MIT jaillet@mit.edu
Pseudocode Yes Algorithm 1: Generic UCB
Open Source Code No The paper does not provide any statement about releasing open-source code for the described methodology, nor does it provide any links to code repositories.
Open Datasets No The paper is theoretical and defines conceptual bandit models (MAB, LCB) but does not refer to or use any specific public or open datasets for empirical training or evaluation.
Dataset Splits No The paper is theoretical and does not conduct empirical experiments, thus no training/validation/test dataset splits are provided.
Hardware Specification No The paper is theoretical and does not describe any hardware used for experiments.
Software Dependencies No The paper is theoretical and does not mention any specific software dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not describe an experimental setup with hyperparameters or system-level training settings. Algorithm 1 is a generic description, not a specific experimental configuration.