Bounding Causal Effects on Continuous Outcome
Authors: Junzhe Zhang, Elias Bareinboim12207-12215
AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, we apply the derived causal bounds to various bandit learning algorithms (Gittins 1979), showing that they could consistently improve the convergence for identifying the optimal treatment. Our results are validated on the International Stroke Trial data (Carolei et al. 1997). We now use a real-world dataset to investigate the performance of proposed bandit strategies. |
| Researcher Affiliation | Academia | Junzhe Zhang, Elias Bareinboim Causal Artiļ¬cial Intelligence Laboratory Columbia University {junzhez,eb}@cs.columbia.edu |
| Pseudocode | Yes | Algorithm 1: Causal-UCB (UCBc) |
| Open Source Code | No | The paper refers to a technical report (Zhang and Bareinboim 2020) at URL https://causalai.net/r61.pdf, but this link is to a PDF document, not an explicit statement of code release or a code repository. |
| Open Datasets | Yes | Our results are validated on the International Stroke Trial data (Carolei et al. 1997). Carolei, A.; et al. 1997. The International Stroke Trial (IST): a randomized trial of aspirin, subcutaneous heparin, both, or neither among 19435 patients with acute ischaemic stroke. The Lancet 349: 1569 1581. |
| Dataset Splits | No | The paper mentions using total sample sizes like 'n = 9650 samples' and 'n = 9711 observations' for estimation and analysis, but does not specify any explicit training, validation, or test dataset splits or their percentages. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions various algorithms and methods (e.g., 'kl-UCB procedure', 'D-UCB', 'standard LP methods') but does not specify any software dependencies or libraries with version numbers. |
| Experiment Setup | No | For details on the experimental setups, we refer readers to the full technical report (Zhang and Bareinboim 2020). The main text does not provide specific hyperparameter values or training configurations. |