اخبار و اعلانات

 آمار

تعداد نشریات49
تعداد شماره‌ها1,282
تعداد مقالات11,081
تعداد مشاهده مقاله22,936,162
تعداد دریافت فایل اصل مقاله15,485,157
Güldoğan Dericioğlu, Yaprak. (1405). Learning-Assisted Adaptive Krylov Solvers for Large-Scale Matrix Differential Riccati Equations. فصلنامه علمی زیست فناوری گیاهان زراعی, (), -. doi: 10.30473/coam.2026.76340.1355
Yaprak Güldoğan Dericioğlu. "Learning-Assisted Adaptive Krylov Solvers for Large-Scale Matrix Differential Riccati Equations". فصلنامه علمی زیست فناوری گیاهان زراعی, , , 1405, -. doi: 10.30473/coam.2026.76340.1355
Güldoğan Dericioğlu, Yaprak. (1405). 'Learning-Assisted Adaptive Krylov Solvers for Large-Scale Matrix Differential Riccati Equations', فصلنامه علمی زیست فناوری گیاهان زراعی, (), pp. -. doi: 10.30473/coam.2026.76340.1355
Güldoğan Dericioğlu, Yaprak. Learning-Assisted Adaptive Krylov Solvers for Large-Scale Matrix Differential Riccati Equations. فصلنامه علمی زیست فناوری گیاهان زراعی, 1405; (): -. doi: 10.30473/coam.2026.76340.1355

Learning-Assisted Adaptive Krylov Solvers for Large-Scale Matrix Differential Riccati Equations

Control and Optimization in Applied Mathematics
مقالات آماده انتشار، پذیرفته شده، انتشار آنلاین از تاریخ 09 اردیبهشت 1405    اصل مقاله (701.87 K)
نوع مقاله: Research Article
شناسه دیجیتال (DOI): 10.30473/coam.2026.76340.1355
نویسنده
Yaprak Güldoğan Dericioğlu*
Bitlis Eren University, Department of Mathematics, Bitlis, Türkiye
چکیده
The Extended Block Arnoldi–Backward Differentiation Formula (EBA–BDF) is a projection-based integrator for solving large-scale Matrix Differential Riccati Equations (DREs). Like many Krylov subspace methods, its performance depends on the choice of the subspace dimension. In practice, this parameter is often determined through empirical tuning. In this work, we introduce a lightweight, data-driven pre-solver to estimate this dimension \emph{a priori}. The approach uses a Random Forest model trained on spectral norms and discretization parameters, and predicts the required subspace size without modifying the numerical core or stability properties of the original method. Numerical experiments show that the proposed approach can automate parameter selection and reduce the need for manual tuning. The effect is more noticeable in diffusion-dominated regimes, where spectral properties lead to more regular Krylov convergence. By simplifying the initialization stage, the approach supports the practical use of EBA–BDF solvers in large-scale problems.

تازه های تحقیق
  • A data-driven pre-solver estimates the Krylov subspace dimension before integration begins.
  • Random Forest regressor trained on spectral norms, conditioning, and discretization parameters.
  • Achieves R² > 0.97 in diffusion-dominated regimes and R² ≈ 0.79 in convection–reaction cases.
  • Reduces EBA–BDF(2) solver runtime by 40–50% without modifying the numerical core.
  • Feature importance rankings align with classical Krylov convergence theory, validating the approach.
کلیدواژه‌ها
Low-rank differential Riccati equation؛ Krylov subspace projection؛ Backward differentiation formula؛ Learning-assisted solvers؛ Learning-assisted numerical solvers؛ Spectral descriptors؛ Machine learning
مراجع

[1] Abou-Kandil, H., Freiling, G., Ionescu, V., Jank, G. (2003). “Matrix Riccati equations in control and systems theory”. Birkhäuser Basel. https://doi.org/10.1007/ 978-3-0348-8081-7

[2] Arias, E., Hernández, V., Ibáñez, J.J., Peinado, J. (2007). “A fixed point-based BDF method for solving differential Riccati equations”. Applied Mathematics and Computation, 188(2), 1319–1333. https://doi.org/10.1016/j.amc.2006.11.001

[3] Benner, P., Mena, H. (2004). “BDF methods for large-scale differential Riccati equations”. Proceedings of the 16th International Symposium on Mathematical Theory of Networks and Systems, 1–12. https://csc.mpi-magdeburg.mpg.de/mpcsc/benner/pub/BM_ mtns04.pdf

[4] Benner, P., Kürschner, P., Saak, J. (2013). “Efficient handling of complex shift parameters in the low-rank Cholesky-factor ADI method”. Numerical Algorithms, 62(2), 225–251. https://doi.org/10.1007/s11075-012-9569-7

[5] Benner, P., Li, J.-R., Penzl, T. (2008). “Numerical solution of large-scale Lyapunov equations, Riccati equations, and linear-quadratic optimal control problems”. Numerical Linear Algebra with Applications, 15(9–10), 755–777. https://doi.org/10.1002/nla.

[6] Botchev, M.A., Grimm, V., Hochbruck, M. (2013). “Residual, restarting, and Richardson iteration for the matrix exponential”. SIAM Journal on Scientific Computing, 35(3), A1376–A1397. https://doi.org/10.1137/110820191

[7] Breiman, L. (2001). “Random forests”. Machine Learning, 45(1), 5–32. https://doi. org/10.1023/A:1010933404324

[8] Chegeni, M.S., Yarahmadi, M. (2025). “Optimal control of linear singularly perturbed systems via eigenvalue assignment”. Control and Optimization in Applied Mathematics, 10(1), 1–17. https://doi.org/10.30473/coam.2025.73999. 1294

[9] Chen, J. (2025). “Graph neural preconditioners for iterative solutions of sparse linear systems”. International Conference on Learning Representations. https://doi.org/10.48550/arXiv.2406.00809

 

[10] Chen, H., Dong, H., Geng, Z., Kuang, Y., Luo, J., Wang, H., Wang, J. (2024). “Neural Krylov iteration for accelerating linear system solving”. Advances in Neural Information Processing Systems, 37, 128636–128667. https://doi.org/10.52202/079017-4086

[11] Dieci, L., Eirola, T. (1994). “Positive definiteness in the numerical solution of Riccati differential equations”. Numerische Mathematik, 67(3), 303–313. https://doi.org/ 10.1007/s002110050030

[12] Druskin, V., Knizhnerman, L.A. (1998). “Extended Krylov subspaces: Approximation of the matrix square root and related functions”. SIAM Journal on Matrix Analysis and Applications, 19, 755–771. https://doi.org/10.1137/S0895479895292400

[13] Druskin, V., Simoncini, V. (2011). “Adaptive rational Krylov subspaces for large-scale dynamical systems”. Systems & Control Letters, 60(8), 546–560. https://doi.org/ 10.1016/j.sysconle.2011.04.013

[14] Güldoğan, Y., Hached, M., Jbilou, K., Kurulay, M. (2018). “Low rank approximate solutions to large-scale differential matrix Riccati equations”. Applicationes Mathematicae, 45(2), 233–254. https://doi.org/10.4064/am2355-1-2018

[15] Hached, M., Jbilou, K., Güldoğan, Y. (2025). “A review on the numerical resolution of large-scale low-rank differential matrix equations”. In: Jbilou, K., Hached, M., Maniar, L., Ratnani, A. (eds) Mathematical Modeling with Modern Applications. M3A 2024. Springer Proceedings in Mathematics & Statistics, vol 497. Springer, Cham, 151–180. https://doi.org/10.1007/978-3-031-89041-3_8

[16] Häusner, P., Öktem, O., Sjölund, J. (2024). “Neural incomplete factorization: Learning preconditioners for the conjugate gradient method”. Transactions on Machine Learning Research. https://doi.org/10.48550/arXiv.2305.16368

[17] Lancaster, P., Rodman, L. (1995). “Algebraic Riccati equations”. Oxford: Clarendon Press / Oxford University Press. https://doi.org/10.1093/oso/9780198537953. 001.0001

[18] Li, Y., Chen, P.Y., Du, T., Matusik, W. (2023). “Learning preconditioners for conjugate gradient PDE solvers”. Proceedings of the 40th International Conference on Machine Learning, PMLR, 202, 19425–19439. https://proceedings.mlr.press/ v202/li23e.html

[19] Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., VanderPlas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., Duchesnay, É. (2011). “Scikit-learn: machine learning in Python”. Journal of Machine Learning Research, 12(85), 2825–2830. http: //jmlr.org/papers/v12/pedregosa11a.html

[20] Saad, Y. (2003). “Iterative methods for sparse linear systems”, 2nd ed., Society for Industrial and Applied Mathematics. https://doi.org/10.1137/1.9780898718003

[21] Simoncini, V. (2016). “Computational methods for linear matrix equations”. SIAM Review, 58(3), 377–441. https://doi.org/10.1137/130912839

آمار
تعداد مشاهده مقاله: 18
تعداد دریافت فایل اصل مقاله: 9


سامانه مدیریت نشریات علمی. قدرت گرفته از سیناوب