Adaptively Exploiting d-Separators with Causal Bandits

Authors: Blair Bilodeau, Linbo Wang, Dan Roy

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Empirically, we observe these performance improvements on simulated data.
Researcher Affiliation Academia Blair Bilodeau University of Toronto Linbo Wang University of Toronto Daniel M. Roy University of Toronto
Pseudocode Yes Algorithm 1: HAC-UCB(A, Z, T, (Z))
Open Source Code Yes Implementation details are available in Appendix D and code can be found at https://github.com/blairbilodeau/adaptive-causal-bandits.
Open Datasets No The paper uses 'simulated data' (Section 5). It does not provide concrete access information (link, DOI, repository, or citation) for a publicly available or open dataset.
Dataset Splits No The paper describes simulated environments but does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) for training, validation, or test sets.
Hardware Specification No The paper states: 'All computations were done on CPU on a personal laptop computer.' This is not specific enough as it does not include exact CPU models, processor types, or memory details.
Software Dependencies No The paper does not provide specific ancillary software details, such as library or solver names with version numbers.
Experiment Setup Yes The paper describes the setup for its simulations in Section 5, including: 'Taking the gap = |A| (log T)/T, the fixed conditional distribution for Z = {0, 1} is Y | Z Ber(1/2 + (1 Z) ). Then, for a small Á (we take Á = 0.0005), we set P 1[Z = 0] = 1 Á and P a[Z = 0] = Á for all other a œ A \ {1}.'