# Tuning-Free Structured Sparse PCA via Deep Unfolding Networks

arXiv: https://arxiv.org/abs/2502.20837 · HTML: https://mlsanity.com/paper/2502.20837/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 39 references that are in the map.
- First posted on arXiv 28 Feb 2025.

## Builds on (most cited first, 10 of 10)

- [Model-Based Deep Learning](https://mlsanity.com/paper/2012.08405/) (2020)
- [Deep Equilibrium Architectures for Inverse Problems in Imaging](https://mlsanity.com/paper/2102.07944/) (2021)
- [Feature Selection: A Data Perspective](https://mlsanity.com/paper/1601.07996/) (2016)
- [Learned reconstruction methods with convergence guarantees](https://mlsanity.com/paper/2206.05431/) (2022)
- [RPCANet: Deep Unfolding RPCA Based Infrared Small Target Detection](https://mlsanity.com/paper/2311.00917/) (2023)
- [A systematic review of Hate Speech automatic detection using Natural Language Processing](https://mlsanity.com/paper/2106.00742/) (2021)
- [Sparse PCA via l2,p-Norm Regularization for Unsupervised Feature Selection](https://mlsanity.com/paper/2012.14595/) (2020)
- [Learning Feature Sparse Principal Components](https://mlsanity.com/paper/1904.10155/) (2019)
- [Learning to optimize: A tutorial for continuous and mixed-integer optimization](https://mlsanity.com/paper/2405.15251/) (2024)
- [Joint Communications and Sensing Hybrid Beamforming Design via Deep Unfolding](https://mlsanity.com/paper/2307.04376/) (2023)

## Abstract (arXiv)

Sparse principal component analysis (PCA) is a well-established dimensionality reduction technique that is often used for unsupervised feature selection (UFS). However, determining the regularization parameters is rather challenging, and conventional approaches, including grid search and Bayesian optimization, not only bring great computational costs but also exhibit high sensitivity. To address these limitations, we first establish a structured sparse PCA formulation by integrating ℓ1-norm and ℓ2,1-norm to capture the local and global structures, respectively. Building upon the off-the-shelf alternating direction method of multipliers (ADMM) optimization framework, we then design an interpretable deep unfolding network that translates iterative optimization steps into trainable neural architectures. This innovation enables automatic learning of the regularization parameters, effectively bypassing the empirical tuning requirements of conventional methods. Numerical experiments on benchmark datasets validate the advantages of our proposed method over the existing state-of-the-art methods. Our code will be accessible at https://github.com/xianchaoxiu/SPCA-Net.