A Near-Optimal Change-Detection Based Algorithm for Piecewise-Stationary Combinatorial Semi-Bandits

Authors: Huozhi Zhou, Lingda Wang, Lav Varshney, Ee-Peng Lim6933-6940

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical experiments on both synthetic and realworld datasets demonstrate the superiority of GLR-CUCB compared to other state-of-the-art algorithms.
Researcher Affiliation Academia 1Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign 2School of Information Systems, Singapore Management University {hzhou35, lingdaw2, varshney}@illinois.edu, eplim@smu.edu.sg
Pseudocode Yes Algorithm 1 Sub-Bernoulli GLR Change-Point Detector: GLR(X1, , Xn; δ)... Algorithm 2 The GLR-CUCB Algorithm
Open Source Code No The paper does not provide a direct link or explicit statement that the source code for their proposed method (GLR-CUCB) is open-source or available. It only mentions that a baseline algorithm's code was provided by another researcher.
Open Datasets Yes We adopt the benchmark dataset for the real-world evaluation of bandit algorithms from Yahoo!. ... Yahoo! Front Page Today Module User Click Log Dataset on https://webscope.sandbox.yahoo.com
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, or detailed splitting methodology) for training, validation, and testing. It mentions using 'synthetic and real-world datasets' but doesn't detail their splits.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiment.
Experiment Setup Yes In the synthetic dataset experiments: 'We let T = 5000, K = 6, m = 2, and N = 5.' In the Yahoo! Dataset experiments: 'To make the experiment nontrivial, we modify the dataset by: 1) the click rate of each base arm is enlarged by 10 times; 2) Reducing the time horizon to T = 22500.' Also, 'The details about parameter tuning for all of these algorithms for different experiments are included in Appendix C.2.'