Contextual Combinatorial Bandits with Probabilistically Triggered Arms

Authors: Xutong Liu, Jinhang Zuo, Siwei Wang, John C.S. Lui, Mohammad Hajiesmaili, Adam Wierman, Wei Chen

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We also include experiments that demonstrate the improved performance of our algorithms compared with benchmark algorithms on synthetic and real-world datasets.
Researcher Affiliation Collaboration 1The Chinese University of Hong Kong, Hong Kong SAR, China 2University of Massachusetts Amherst, MA, United States 3California Institute of Technology, CA, United States 4Microsoft Research, Beijing, China.
Pseudocode Yes Algorithm 1 C2-UCB-T: Contextual Combinatorial Upper Confidence Bound Algorithm for C2MAB-T
Open Source Code No No explicit statement about the release of their own source code or a link to a code repository was found.
Open Datasets Yes Movie Lens-1M dataset grouplens.org/datasets/movielens/1m/
Dataset Splits No No explicit mention of specific train/validation/test dataset splits (e.g., percentages or sample counts) was found. The paper describes data usage but not data partitioning for training, validation, or testing.
Hardware Specification No No specific hardware details (e.g., CPU, GPU models, or memory specifications) used for running the experiments were mentioned in the paper.
Software Dependencies No No specific software dependencies with version numbers were mentioned in the paper (e.g., programming languages or libraries like Python, PyTorch, or scikit-learn).
Experiment Setup Yes We set d = 20, K = 4, and the goal is to choose K out of m movies to maximize the reward of the cascading recommendation. We use their learned feature mapping ϕ from movies to the probability that a uniformly random user rated the movie more than three stars.