Stochastic Graphical Bandits with Adversarial Corruptions

Authors: Shiyin Lu, Guanghui Wang, Lijun Zhang8749-8757

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The effectiveness of our algorithm is demonstrated by numerical experiments.
Researcher Affiliation Academia National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210023, China {lusy, wanggh, zhanglj}@lamda.nju.edu.cn
Pseudocode Yes Algorithm 1 Elise
Open Source Code No The paper does not provide an explicit statement about releasing source code for the described methodology or a direct link to a code repository.
Open Datasets No The paper describes how the data for experiments is generated synthetically (e.g., 'generate the stochastic rewards', 'Erdos Renyi model to generate the feedback graph') rather than utilizing or providing access to a pre-existing public dataset.
Dataset Splits No The paper describes a sequential decision-making process (bandits) and does not use traditional dataset splits like training, validation, and test sets.
Hardware Specification No The paper does not provide any specific details about the hardware used to run the experiments.
Software Dependencies No The paper does not specify any software dependencies with version numbers.
Experiment Setup Yes We use a relatively large time horizon T = 2000000 and set the corruption level as C = 1000 ln(T). We set K = 10 and adopt the Erdos Renyi model to generate the feedback graph... Specifically, for each pair of arms (u, v) [K] [K] with u = v, we connect them with a fixed probability p.