|
|
Pure Mathematics 2024
一类带变换罚函数的正则化问题的强度量次正则研究
|
Abstract:
本文研究一类带变换？1罚函数的正则化问题，该模型的目标函数由损失函数和变换？1罚函数两部分组成，其中损失函数是二次可微函数，变换？1罚函数是一个非凸函数。本文首先研究变换？1罚函数的邻近次微分和极限次微分，然后利用变换？1罚函数的次微分表达式和集值映射的图像导数工具得到了该类问题目标函数的次微分的图像导数，最后利用该图像导数表达式分别建立了该正则化问题的强度量次正则的一个充分条件和充要条件。
This paper studies the regularization problem for a class of penalty functions with transformed？1. The objective function of the model consists of two parts: the loss function and the transformed？1penalty function, where the loss function is a quadratic differentiable function and the transformed？1penalty function is a nonconvex function. This article first studies the proximal subdifferentials and limiting subdifferentials of the transformed？1penalty function. Then, by the subdifferential expression of the transformed？1penalty function and graphical derivative tool for set-value mapping, we obtain the graphical derivative of the subdifferentials of the objective function. Finally, using the graphical derivative expression, we establish a sufficient condition, a necessary and sufficient condition of the strong metric subregularity for the regularization problem.
| [1] | Nikolova, M. (2001) Local Strong Homogeneity of a Regularized Estimator. SIAM Journal on Applied Mathematics, 61, 633-658. https://doi.org/10.1137/S0036139997327794 |
| [2] | Donoho, D.L. (2006) Compressed Sensing. IEEE Transactions on Information Theory, 52, 1289-1306. https://doi.org/10.1109/TIT.2006.871582 |
| [3] | Fan, J. and Li, R. (2001) Variable Selection via Nonconcave Penalized Likelihood and Its Oracle Properties. Journal of the American Statistical Association, 96, 1348-1360. https://doi.org/10.1198/016214501753382273 |
| [4] | Lv, J. and Fan, Y. (2009) A Unified Approach to Model Selection and Sparse Recovery Using Regularized Least Squares. The Annals of Statistics, 37, 3498-3528. https://doi.org/10.1214/09-AOS683 |
| [5] | Zhang, S. and Xin, J. (2014) Minimization of Transformed Penalty: Theory, Difference of Convex Function Algorithm, and Robust Application in Compressed Sensing. Mathmatical Programming, 169, 307-336. https://doi.org/10.1007/s10107-018-1236-x |
| [6] | Zhang, S. and Xin, J. (2016) Minimization of Transformed Penalty: Closed Form Representation and Iterative Thresholding Algorithms. Communications in Mathematical Sciences, 15, 511-537. https://doi.org/10.4310/CMS.2017.v15.n2.a9 |
| [7] | Ahn, M., Pang, J.-S. and Xin, J. (2017) Difference-of-Convex Learning: Directional Stationarity, Optimality, and Sparsity. SIAM Journal on Optimization, 27, 1637-1665. https://doi.org/10.1137/16M1084754 |
| [8] | Zhang, S., Yin, P.H. and Xin, J. (2017) Transformed Schatten-1 Iterative Thresholding Algorithms for Low Rank Matrix Completion. Communications in Mathematical Sciences, 15, 839-862. https://doi.org/10.4310/CMS.2017.v15.n3.a12 |
| [9] | Guo, L., Li, J. and Liu, Y. (2018) Stochastic Collocation Methods via Minimisation of the Transformed-Penalty. East Asian Journal on Applied Mathematics, 8, 566-585. https://doi.org/10.4208/eajam.060518.130618 |
| [10] | Ma, R.R., Miao, J.Y., Niu, L.F., et al. (2019) Transformed Regularization for Learning Sparse Deep Neural Networks. Neural Networks, 119, 286-298. https://doi.org/10.1016/j.neunet.2019.08.015 |
| [11] | Bello-Cruz, Y., Li, G.Y. and Nghia, T.T.A. (2022) Quadratic Growth Conditions and Uniqueness of Optimal Solution to Lasso. Journal of Optimization Theory and Applications, 194, 167-190. https://doi.org/10.1007/s10957-022-02013-2 |
| [12] | Li, M.H., Meng, K.W. and Yang, X.Q. (2023) Variational Analysis of Kurdyka-？ojasiewicz Property, Exponent and Modulus. https://arxiv.org/abs/2308.15760v2 |
| [13] | Chieu, N.H. and Hien, L.V. (2017) Computation of Graphical Derivative for a Class of Normal Cone Mappings under a Very Weak Condition. SIAM Journal on Optimization, 27, 190-204. https://doi.org/10.1137/16M1066816 |
| [14] | Gfrerer, H. and Mordukhovich, B.S. (2019) Second-Order Variational Analysis of Parametric Constraint and Variational Systems. SIAM Journal on Optimization, 29, 423-453. https://doi.org/10.1137/17M1157751 |
| [15] | Henrion, R., Kruger, A.Y. and Outrata, J.V. (2013) Some Remarks on Stability of Generalized Equations. Journal of Optimization Theory and Applications, 159, 681-697. https://doi.org/10.1007/s10957-012-0147-x |
| [16] | Dontchev, A.L. and Rockafellar, R.T. (2009) Implicit Functions and Solution Mappings. A View from Variational Analysis. Springer, New York. https://doi.org/10.1007/978-0-387-87821-8 |
| [17] | Zheng, X.Y. and Ng, K.F. (2007) Metric Subregularity and Constraint Qualifications for Convex Generalized Equations in Banach Spaces. SIAM Journal on Optimization, 18, 437-460. https://doi.org/10.1137/050648079 |
| [18] | Chieu, N.H., Hien, L.V., Nghia, T.T.A., et al. (2021) Quadratic Growth and Strong Metric Subregularity of the Subdifferential via Subgradient Graphical Derivative. SIAM Journal on Optimization, 31, 545-568. https://doi.org/10.1137/19M1242732 |
| [19] | Cibulka, R., Dontchev, A.L. and Kruger, A.Y. (2017) Strong Metric Subregularity of Mappings in Variational Analysis and Optimization. Journal of Mathematical Analysis and Applications, 457, 1247-1282. https://doi.org/10.1016/j.jmaa.2016.11.045 |
| [20] | Aragón Artacho, F.J. and Geoffroy, M.H. (2013) Metric Subregularity of the Convex Subdifferential in Banach Spaces. Journal of Nonlinear and Convex Analysis, 15, 35-47. |
| [21] | Drusvyatskiy, D., Mordukhovich, B.S. and Nghia, T.T.A. (2014) Second-Order Growth, Tilt Stability, and Metric Regularity of the Subdifferential. Journal of Convex Analysis, 21, 1165-1192. |
| [22] | Chieu, N.H., Trang, N.T.Q. and Tuan, H.A. (2022) Quadratic Growth and Strong Metric Subregularity of the Subdifferential for a Class of Non-Prox-Regular Functions. Journal of Optimization Theory and Applications, 194, 1081-1106. https://doi.org/10.1007/s10957-022-02071-6 |
| [23] | Clarke, F.H., Ledyaev, Y.S., Stern, R.J., et al. (1998) Nonsmooth Analysis and Control Theory. Springer, New York. |
| [24] | Rockafellar, R.T. and Wets, R.J.-B. (2009) Variational Analysis. Springer, Heidelberg. |
