Nonstochastic Multiarmed Bandits with Unrestricted Delays

Authors: Tobias Sommer Thune, Nicolò Cesa-Bianchi, Yevgeny Seldin

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We first prove that "delayed" Exp3 achieves the O p (KT + D) ln K regret bound conjectured by Cesa-Bianchi et al. [2019] in the case of variable, but bounded delays. ... This section contains the main points of the analysis of Algorithm 1 leading to the proof of Theorem 1...
Researcher Affiliation Academia Tobias Sommer Thune University of Copenhagen Copenhagen, Denmark... Nicolò Cesa-Bianchi DSRC & Univ. degli Studi di Milano Milan, Italy... Yevgeny Seldin University of Copenhagen Copenhagen, Denmark
Pseudocode Yes Algorithm 1: Delayed exponential weights (DEW) Input : Learning rate η; upper bound on the delays dmax... Algorithm 2: Skipper Input : Threshold β; Algorithm A.
Open Source Code No The paper does not provide any links to source code repositories or explicit statements about the availability of open-source code for the methodology described.
Open Datasets No This is a theoretical paper focusing on algorithms and regret bounds for multiarmed bandits. It does not utilize or describe any datasets for training, therefore no information about public availability of datasets is provided.
Dataset Splits No This is a theoretical paper that does not involve empirical data or experiments. As such, there is no mention of training, validation, or test dataset splits.
Hardware Specification No As a theoretical paper, it focuses on algorithms and mathematical proofs rather than empirical evaluation. Therefore, no specific hardware specifications used for experiments are mentioned.
Software Dependencies No The paper describes algorithms and their theoretical properties. It does not mention any specific software dependencies or their version numbers that would be required to reproduce experiments.
Experiment Setup No This paper is theoretical and focuses on algorithm design and analysis, including parameters for the algorithms like learning rates and thresholds. However, it does not provide details of an empirical experimental setup such as hyperparameters for training models on data or system-level settings, as no such experiments are described.