Identification of Linear Non-Gaussian Latent Hierarchical Structure
Authors: Feng Xie, Biwei Huang, Zhengming Chen, Yangbo He, Zhi Geng, Kun Zhang
ICML 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on both synthetic and real-world data show the effectiveness of the proposed approach. In this section, we first apply the proposed method to synthetic data to demonstrate the correctness. Then, we apply our algorithm to real-world data set to show its usefulness. |
| Researcher Affiliation | Academia | 1Department of Probability and Statistics, Peking University, Beijing, China 2Department of Applied Statistics, Beijing Technology and Business University, Beijing, China 3Department of Philosophy, Carnegie Mellon University, Pittsburgh, USA 4School of Computer Science, Guangdong University of Technology, Guangzhou, China 5Machine Learning Department, Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, UAE. |
| Pseudocode | Yes | The complete procedure is summarized in Algorithm 1, and an illustrative example for each phase will be given immediately after introducing each phase in the following subsections, and a complete example is given in Appendix D. (Algorithm 1 is presented on page 4). |
| Open Source Code | Yes | The source code is in the Supplementary file. |
| Open Datasets | Yes | We applied our La HME algorithm to a multitasking behavior model, represented by a hierarchical SEM (Himi et al., 2019). The detailed explanation of the data set is given in Appendix F. (Appendix F cites Himi et al., 2019 again). |
| Dataset Splits | No | The paper mentions 'different sample sizes N = 1k, 5k, 10k' for synthetic data and '202 samples' for real-world data, but it does not specify explicit training/validation/test dataset splits (e.g., percentages or counts). |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running experiments, such as GPU/CPU models or memory. |
| Software Dependencies | No | The paper mentions using 'HSIC-based independence tests (Zhang et al., 2018)' but does not provide specific version numbers for this or any other software dependencies. |
| Experiment Setup | No | The paper describes how synthetic data was generated ('causal strength... generated uniformly', 'non-Gaussian noise terms... from exponential distributions'), but it does not provide specific hyperparameters or system-level training settings for the La HME algorithm itself. |