Alternating minimization algorithm for unlabeled sensing and linked linear regression

This repository contains the MATLAB implementation of the alternating minimization algorithm proposed in the paper Alternating Minimization algorithm for unlabeled sensing and linked linear regression.

DOI: https://doi.org/10.1016/j.sigpro.2025.109927

Repository Structure

The repository has 6 foldes: "Fig3", "Table1", "altMinProposed", "benchmarks", "dataSets", "misc". Below, I describe each of these folders and the files they contain.

• Fig3: To reproduce the results in Figure 3 of the paper. For example, to reproduce Fig. 3(a), run the fig3a.m file in the folder Fig3. The Python code for the stochastic EM benchmark algorithm in this folder was taken from the authors' own implementation at https://github.com/abidlabs/stochastic-em-shuffled-regression/tree/master,
• Table1: To reproduce the results in Table 1 of the paper.
• altMinProposed: Code for the implementation of the proposed Alternating Minimization algorithm/
• benchmarks: Code for the benchmark methods proposed in other works.
• dataSets: The real datasets used for the results in Table 1.
• misc: Commonly used function throughout the code.

Instructions

• For linux systems, replace ' \ ' at the beginning of the the main.m files in the figures folder by '/'.
• To reproduce any of the results in [1], run the corerresponding file in the Fig3 folder. For example, to reproduce Fig. 3(a), run the fig3a.m file in the folder Fig3.
• To reproduce the results on real-datasets in Table 1, please first install CVX.
• The sarcos dataset (see Table 1 of Reference) is a large dataset, with a long run-time. The results for that dataset are provided in the MATLAB data file sarcos.mat

References

[1]. Alternating Minimization algorithm for unlabeled sensing and linked linear regression. URL

[2]. Ahmed Ali Abbasi, Abiy Tasissa, and Shuchin Aeron. R-local unlabeled sensing: A novel graph matching approach for multiview unlabeled sensing under local permutations. IEEE Open Journal of Signal Processing, 2:309–317, 2021. URL

Feedback

Please email any feedback to the author Ahmed Ali Abbasi at aabbasi1@iastate.edu.

Citation

Please cite the following papers on Unlabeled Sensing if you use this code and if these papers are relevant to your paper. Thank you!

1. @ARTICLE{9440727,
  author={Abbasi, Ahmed Ali and Tasissa, Abiy and Aeron, Shuchin},
  journal={IEEE Open Journal of Signal Processing}, 
  title={R-Local Unlabeled Sensing: A Novel Graph Matching Approach for Multiview Unlabeled Sensing Under Local Permutations}, 
  year={2021},
  volume={2},
  number={},
  pages={309-317},
  keywords={Sensors;Loss measurement;Covariance matrices;Signal processing algorithms;Signal processing;Task analysis;Minimization;Graph matching;multi-view;unlabeled sensing},
  doi={10.1109/OJSP.2021.3083479}}

2. @INPROCEEDINGS{9746201,
  author={Abbasi, Ahmed Ali and Tasissa, Abiy and Aeron, Shuchin},
  booktitle={ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, 
  title={r-Local Unlabeled Sensing: Improved Algorithm and Applications}, 
  year={2022},
  volume={},
  number={},
  pages={5593-5597},
  keywords={Geometry;Linear systems;Signal processing algorithms;Signal processing;Minimization;Mathematical models;Sensors;Unlabeled sensing;unassigned distance geometry problem;proximal alternating minimization},
  doi={10.1109/ICASSP43922.2022.9746201}}

3.@article{ABBASI2025109927,
title = {Alternating minimization algorithm for unlabeled sensing and linked linear regression},
journal = {Signal Processing},
volume = {232},
pages = {109927},
year = {2025},
issn = {0165-1684},
doi = {https://doi.org/10.1016/j.sigpro.2025.109927},
url = {https://www.sciencedirect.com/science/article/pii/S0165168425000428},
author = {Ahmed Ali Abbasi and Shuchin Aeron and Abiy Tasissa}}