Dynamic Learning of Sequential Choice Bandit Problem under Marketing Fatigue

Authors: Junyu Cao, Wei Sun3264-3271

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

Reproducibility Variable Result LLM Response
Research Type Experimental 7 Numerical Experiments In this section, we first investigate the robustness of Algorithm 1 which is our proposed UCB-algorithm for the SC-Bandit problem by comparing how the regret changes with respect to different values of u. Next, we compare our Algorithm 1 and 2 with two benchmarks in the non-contextual and contextual settings respectively.
Researcher Affiliation Collaboration Junyu Cao University of California, Berkeley Berkeley, California 94720 jycao@berkeley.edu Wei Sun IBM Research Yorktown Height, New York 10591 sunw@us.ibm.com
Pseudocode Yes Algorithm 1: An exploration-exploitation algorithm for SC-Bandit under marketing fatigue
Open Source Code No The paper does not provide any concrete access to source code, such as a specific repository link, an explicit code release statement, or code in supplementary materials.
Open Datasets No Experiment setup We consider a setting with N = 30, revenue ri is uniformly distributed between [0,1], abandonment distribution probability p = 0.1 and the cost of abandonment c = 0.5. We present four scenarios, when the valuation u is uniformly generated from [0,0.1], [0,0.2], [0.0.3], and [0.0.5], respectively.
Dataset Splits No The paper describes generating data for experiments and running simulations over a time horizon T, but it does not specify explicit training, validation, or test dataset splits with percentages or counts.
Hardware Specification No The paper does not provide any specific hardware details such as 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, that would be needed to replicate the experiment.
Experiment Setup Yes Experiment setup We consider a setting with N = 30, revenue ri is uniformly distributed between [0,1], abandonment distribution probability p = 0.1 and the cost of abandonment c = 0.5. We present four scenarios, when the valuation u is uniformly generated from [0,0.1], [0,0.2], [0.0.3], and [0.0.5], respectively.