Counterfactual Cross-Validation: Stable Model Selection Procedure for Causal Inference Models
Authors: Yuta Saito, Shota Yasui
ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Empirical evaluations demonstrate that our metric outperforms existing metrics in both model selection and hyperparameter tuning tasks. |
| Researcher Affiliation | Collaboration | 1Tokyo Institute of Technology, 2Cyber Agent, Inc. |
| Pseudocode | Yes | Algorithm 1 Counterfactual Cross-Validation (CF-CV) |
| Open Source Code | Yes | Our code used to conduct the semi-synthetic experiments is available at https://github.com/usaito/counterfactual-cv |
| Open Datasets | Yes | We used the Infant Health Development Program (IHDP) dataset provided by (Hill, 2011). |
| Dataset Splits | Yes | We conducted the experimental procedure over 100 different realizations with 35/35/30 train/validation/test splits. |
| 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 using 'scikit-learn' and 'Econ ML' but does not provide specific version numbers for these or other software dependencies. |
| Experiment Setup | No | The paper mentions tuning hyperparameters and discusses the hyperparameter search space, but it does not provide specific hyperparameter values (e.g., learning rate, batch size, number of epochs) within the main text. |