Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
HMLasso: Lasso with High Missing Rate
Authors: Masaaki Takada, Hironori Fujisawa, Takeichiro Nishikawa
IJCAI 2019 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We theoretically and experimentally show that our proposed method is highly effective even when there are many missing values. ... We demonstrate the effectiveness of our method through both numerical simulations and real-world data experiments. |
| Researcher Affiliation | Collaboration | Masaaki Takada1 , Hironori Fujisawa2 and Takeichiro Nishikawa1 1Toshiba Corporation 2The Institute of Statistical Mathematics |
| Pseudocode | Yes | Algorithm 1 Covariance Estimation with ADMM |
| Open Source Code | No | The paper provides a link to the arXiv preprint of the paper itself (https://arxiv.org/abs/1811.00255) but does not provide an explicit statement or link to the source code for the methodology. |
| Open Datasets | Yes | We evaluated the performance using a real-world residential building dataset [Rafiei and Adeli, 2016] from the UCI datasets repository2. https://archive.ics.uci.edu/ml/datasets/Residential+Building+Data+Set |
| Dataset Splits | Yes | We randomly split data into 300 samples for training, 36 for validation, and 36 for testing, and iterated the experiments 30 times. |
| Hardware Specification | No | The paper mentions running numerical experiments but does not provide any specific details about the hardware used (e.g., GPU/CPU models, memory). |
| Software Dependencies | No | The paper discusses various algorithms and methods (e.g., coordinate descent, ADMM, mice, missForest) but does not specify any software libraries or dependencies with their version numbers that were used for implementation. |
| Experiment Setup | Yes | The regularization parameter λ was selected by five-fold corrected cross-validation as with [Datta and Zou, 2017]. |