Estimating the Causal Effect from Partially Observed Time Series
Authors: Akane Iseki, Yusuke Mukuta, Yoshitaka Ushiku, Tatsuya Harada3919-3926
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments based on synthesized and real data demonstrate the ability of the proposed method to estimate causal relationships more correctly than existing methods when the data contain missing values, the dimensionality is large, and the number of samples is small. |
| Researcher Affiliation | Academia | Akane Iseki,1 Yusuke Mukuta,1,2 Yoshitaka Ushiku,1 Tatsuya Harada1,2 1The University of Tokyo 2RIKEN AIP {iseki, mukuta, ushiku, harada}@mi.t.u-tokyo.ac.jp |
| Pseudocode | No | The paper describes the proposed method mathematically but does not include any explicit pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any specific links or statements about the availability of open-source code for the described methodology. |
| Open Datasets | Yes | We evaluated the performance of the estimation of climatic information flow using the Global Summary of the Day, the meteorological dataset that contains information about climatic element observed by the National Climatic Data Center. and We used the Crowd Segmentation Dataset (Ali and Shah 2007), which contains video clips of crowded scenes and is available on the Internet. |
| Dataset Splits | No | The paper describes methods for generating synthetic data and introducing missing values, and for real-world datasets, it describes controlling the missing ratio (e.g., 'randomly remove 20% or 40% of the sample size'). However, it does not specify explicit train, validation, or test dataset splits with percentages or sample counts. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., CPU/GPU models, memory, or processing units) used for running its experiments. |
| Software Dependencies | No | The paper mentions using the L-BFGS method and the VGG-16 model, but it does not specify any software dependencies with version numbers (e.g., programming languages, libraries, or frameworks). |
| Experiment Setup | Yes | We set the dimension of the latent variable z to 5. (for synthesized data) and We set the dimensionality of the latent variable to 10. (for meteorological data) and We first resized the input frame to 224 224 and then extracted the output of the second pooling layer of VGG-16. |