Small-Disturbance Input-to-State Stability of Perturbed Gradient Flows: Applications to LQR Problem
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913316349149184 |
|---|---|
| author | Cui, Leilei Jiang, Zhong-Ping Sontag, Eduardo D. |
| author_facet | Cui, Leilei Jiang, Zhong-Ping Sontag, Eduardo D. |
| contents | This paper studies the effect of perturbations on the gradient flow of a general nonlinear programming problem, where the perturbation may arise from inaccurate gradient estimation in the setting of data-driven optimization. Under suitable conditions on the objective function, the perturbed gradient flow is shown to be small-disturbance input-to-state stable (ISS), which implies that, in the presence of a small-enough perturbation, the trajectories of the perturbed gradient flow must eventually enter a small neighborhood of the optimum. This work was motivated by the question of robustness of direct methods for the linear quadratic regulator problem, and specifically the analysis of the effect of perturbations caused by gradient estimation or round-off errors in policy optimization. We show small-disturbance ISS for three of the most common optimization algorithms: standard gradient flow, natural gradient flow, and Newton gradient flow. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_02930 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Small-Disturbance Input-to-State Stability of Perturbed Gradient Flows: Applications to LQR Problem Cui, Leilei Jiang, Zhong-Ping Sontag, Eduardo D. Optimization and Control Systems and Control This paper studies the effect of perturbations on the gradient flow of a general nonlinear programming problem, where the perturbation may arise from inaccurate gradient estimation in the setting of data-driven optimization. Under suitable conditions on the objective function, the perturbed gradient flow is shown to be small-disturbance input-to-state stable (ISS), which implies that, in the presence of a small-enough perturbation, the trajectories of the perturbed gradient flow must eventually enter a small neighborhood of the optimum. This work was motivated by the question of robustness of direct methods for the linear quadratic regulator problem, and specifically the analysis of the effect of perturbations caused by gradient estimation or round-off errors in policy optimization. We show small-disturbance ISS for three of the most common optimization algorithms: standard gradient flow, natural gradient flow, and Newton gradient flow. |
| title | Small-Disturbance Input-to-State Stability of Perturbed Gradient Flows: Applications to LQR Problem |
| topic | Optimization and Control Systems and Control |
| url | https://arxiv.org/abs/2310.02930 |