Improved Algorithm on Online Clustering of Bandits

Authors: Shuai Li, Wei Chen, Shuai Li, Kwong-Sak Leung

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

Reproducibility Variable Result LLM Response
Research Type Experimental The experiments on both synthetic and real datasets consistently show the advantage of the new algorithm over existing methods.
Researcher Affiliation Collaboration 1The Chinese University of Hong Kong 2Microsoft 3University of Cambridge
Pseudocode Yes The pseudocode is provided in Algorithm 1.
Open Source Code No The paper does not contain an explicit statement that the source code for the methodology is available, nor does it provide a link to a code repository.
Open Datasets Yes We use the 20m Movie Lens dataset [Harper and Konstan, 2016] which contains 20 million ratings for 2.7 x 10^4 movies by 1.38 x 10^5 users and Yelp dataset2 which contains 4.7 million ratings of 1.57 x 10^5 restaurants from 1.18 million users. The Yelp dataset is more sparse than Movie Lens. For each of the two real datasets, we extract 10^3 items with most ratings and nu = 10^3 users who rate most.
Dataset Splits No The paper describes using synthetic and real datasets, but it does not specify explicit training, validation, or test dataset splits (e.g., percentages or sample counts). The experiments are conducted in an online learning setting where data arrives over time.
Hardware Specification No The paper does not provide any specific details about the hardware used to run the experiments, such as GPU/CPU models or cloud resources.
Software Dependencies No The paper does not provide specific details about ancillary software, such as library names with version numbers, needed to replicate the experiment.
Experiment Setup No The paper states 'The parameters in all algorithms take theoretical values' and 'We do not optimize the parameters of any algorithm and just use theoretical values of the parameters in all algorithms including SCLUB.' While it provides formulas for parameters in Theorem 1 (e.g., 'αθ = 4R p d/λx, αp = 2 and β = R p d ln(1 + T/d) + 2 ln(4mnu)'), it does not provide concrete numerical values for hyperparameters or other system-level training settings directly in the text that would allow direct reproduction without additional calculations or assumptions about unknown variables.