A Change-Detection Based Framework for Piecewise-Stationary Multi-Armed Bandit Problem

Authors: Fang Liu, Joohyun Lee, Ness Shroff

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The performance of the proposed and existing policies are validated by both synthetic and real world datasets, and we show that our proposed algorithms are superior to other existing policies in terms of regret.
Researcher Affiliation Academia Fang Liu, Joohyun Lee, Ness Shroff The Ohio State University Columbus, Ohio 43210 {liu.3977, lee.7119, shroff.11}@osu.edu
Pseudocode Yes Algorithm 1 CD-UCB, Algorithm 2 Two-sided CUSUM
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets Yes Yahoo! has published a benchmark dataset for the evaluation of bandit algorithms (Yahoo! ). The dataset is the user click log for news articles displayed on the Yahoo! Front Page (Li et al. 2011). ... Yahoo! Webscope program. http://webscope.sandbox.yahoo.com/catalog.php?datatype=r&did=49. [Online; accessed 18-Oct-2016].
Dataset Splits No The paper discusses data preparation but does not provide specific training/test/validation dataset splits.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, processor types) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers needed to replicate the experiment.
Experiment Setup Yes In the simulation, the parameters h and α are tuned around h = log(T/γT ) and α = γT /T log(T/γT ) based on the flipping environment. ... Parameters are listed in Table 2.