Estimating Causal Effects Identifiable from a Combination of Observations and Experiments
Authors: Yonghan Jung, Ivan Diaz, Jin Tian, Elias Bareinboim
NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Simulation results corroborate the theory. In this section, we demonstrate the MR-g ID estimator from Definition (4) through Examples (1,2) and Project STAR dataset[Krueger and Whitmore, 2001, Schanzenbach, 2006]. |
| Researcher Affiliation | Academia | 1Purdue University jung222@purdue.edu 2New York University ivan.diaz@nyu.edu 3Iowa State University jtian@iastate.edu 4Columbia University eb@cs.columbia.edu |
| Pseudocode | Yes | Algorithm 1: GID (x, y, Z, P, G) |
| Open Source Code | No | The paper does not contain an explicit statement or link indicating that the source code for the described methodology is publicly available. |
| Open Datasets | Yes | Project STAR dataset[Krueger and Whitmore, 2001, Schanzenbach, 2006]. |
| Dataset Splits | Yes | 1. Randomly partition Dzi into {Dzi,ℓ}ℓ [L]; i.e., Dzi = L ℓ=1Dzi,ℓ, Zi Z and zi DZi. 2. For each fold ℓ [L], let µi+1 ℓ denote learned µi+1 0 using Dzi+1\Dzi+1,ℓfor i = m, , 2; and πi ℓlearned πi 0 for i = 1, , m 1. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU or CPU models, memory, or cloud instance types used for running experiments. |
| Software Dependencies | No | The paper does not specify version numbers for XGBoost or Python, preventing a reproducible description of ancillary software. It mentions: 'We used the XGBoost [Chen and Guestrin, 2016] as a model for estimating nuisances µ, π, {µi}m i=2, {πi}m i=1. We implemented the model using Python.' and refers to a link for 'Detailed parametrization of parameters including learning rates, maximum depth of the trees, etc.' |
| Experiment Setup | No | The paper describes experimental scenarios and how data was simulated, but it does not provide specific hyperparameter values for the XGBoost models used to estimate nuisances, stating only that 'default parameter settings' were used. |