A Note on Entrywise Consistency for Mixed-data Matrix Completion

Authors: Yunxiao Chen, Xiaoou Li

JMLR 2024 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The proposed methods are evaluated by simulation studies and real-data applications for collaborative filtering and large-scale educational assessment.
Researcher Affiliation Academia Yunxiao Chen EMAIL Department of Statistics London School of Economics and Political Science London WC2A 2AE, UK; Xiaoou Li EMAIL School of Statistics University of Minnesota Minneapolis, MN 55455
Pseudocode Yes Algorithm 1: Refinement Procedure without Data Splitting; Algorithm 2: Refinement Procedure with Data Splitting; Algorithm 3: Refinement Procedure with Multiple Data Splittings
Open Source Code Yes The computation code used in Sections 5 and 6 can be found at https://github.com/yunxiaochen/Matrix Completion_Mixed Data.
Open Datasets Yes We apply the proposed method to a Movie Lens dataset for movie recommendation (Harper and Konstan, 2015). We apply the proposed method to data from the 2018 Program for International Student Assessment (PISA; OECD, 2019a)
Dataset Splits Yes To evaluate the procedures, we split the data into training and test datasets, where the training and test sets contain 80% and 20% of the observed entries, respectively. Similar to the above analysis, we split 80% and 20% of the data into training and test sets and evaluate the prediction accuracy by the test-set log-likelihood.
Hardware Specification No The paper does not explicitly describe the hardware used for running its experiments or simulations.
Software Dependencies No The paper states that computation code is available via a GitHub link, but it does not specify any software dependencies with version numbers in the text.
Experiment Setup Yes In the implementation, we set C2 = 2 p r/p in Algorithms 1, 2, and 3. We set ρ = r in the NBE and C = r in the CJMLE.