The week's pick
Random Articles
Reseach Article
The Critical Role of Hyperparameter Tuning in Machine Learning: A Focus on the SVD Method for Matrix Completion
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 187 - Number 24 |
| Year of Publication: 2025 |
| Authors: Faezeh Aghamohammadi, Fatemeh Shakeri |
10.5120/ijca2025925371
|
Faezeh Aghamohammadi, Fatemeh Shakeri . The Critical Role of Hyperparameter Tuning in Machine Learning: A Focus on the SVD Method for Matrix Completion. International Journal of Computer Applications. 187, 24 ( Jul 2025), 1-6. DOI=10.5120/ijca2025925371
Abstract
In machine learning, the determination of hyperparameters plays an essential role. The significant impact of these parameters on the accuracy of algorithms across problem-solving scenarios cannot be denied. Improper selection of values can significantly increase errors and affects outcomes. Low rank matrix completion, an optimization problem to recover and complete a partial matrix, is an example of dealing with hyperparameter tuning. Based on the experimental knowledge, we find that establishing values for hyperparameters is imperative to achieve an optimal solution to this problem. This study investigates the hyperparameter determination of the singular value thresholding (SVT) method and proposes an approach for selecting these parameters to attain superior solutions.
References
- Jian-Feng Cai, Emmanuel J Cand`es, and Zuowei Shen. A singular value thresholding algorithm for matrix completion. SIAM Journal on optimization, 20(4):1956–1982, 2010.
- Emmanuel Candes and Benjamin Recht. Exact matrix completion via convex optimization. Communications of the ACM, 55(6):111–119, 2012.
- Maryam Fazel. Matrix rank minimization with applications. PhD thesis, PhD thesis, Stanford University, 2002.
- Massimo Fornasier, Holger Rauhut, and Rachel Ward. Lowrank matrix recovery via iteratively reweighted least squares minimization. SIAM Journal on Optimization, 21(4):1614– 1640, 2011.
- Jincai Ha, Chuan Li, Xiaohu Luo, and Zhi Wang. Matrix completion via modified schatten 2/3-norm. EURASIP Journal on Advances in Signal Processing, 2023(1):62, 2023.
- Haifeng Wang, Jinchi Chen, and Ke Wei. Implicit regularization and entrywise convergence of riemannian optimization for low tucker-rank tensor completion. Journal of Machine Learning Research, 24(347):1–84, 2023.
- You-WeiWen, Kexin Li, and Hefeng Chen. Accelerated matrix completion algorithm using continuation strategy and randomized svd. Journal of Computational and Applied Mathematics, 429:115215, 2023.
- Quan Yu, Xinzhen Zhang, Yannan Chen, and Liqun Qi. Low tucker rank tensor completion using a symmetric block coordinate descent method. Numerical Linear Algebra with Applications, 30(3):e2464, 2023.
Index Terms
Keywords