Adversarial Reinforcement Learning: A Duality-Based Approach To Solving Optimal Control Problems
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916820685946880 |
|---|---|
| author | Chen, Nan Liu, Mengzhou Wang, Xiaoyan Zhang, Nanyi |
| author_facet | Chen, Nan Liu, Mengzhou Wang, Xiaoyan Zhang, Nanyi |
| contents | We propose an adversarial deep reinforcement learning (ADRL) algorithm for high-dimensional stochastic control problems. Inspired by the information relaxation duality, ADRL reformulates the control problem as a min-max optimization between policies and adversarial penalties, enforcing non-anticipativity while preserving optimality. Numerical experiments demonstrate ADRL's superior performance to yield tight dual gaps. Our results highlight the potential of ADRL as a robust computational framework for high-dimensional stochastic control in simulation-based optimization contexts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_00801 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adversarial Reinforcement Learning: A Duality-Based Approach To Solving Optimal Control Problems Chen, Nan Liu, Mengzhou Wang, Xiaoyan Zhang, Nanyi Optimization and Control We propose an adversarial deep reinforcement learning (ADRL) algorithm for high-dimensional stochastic control problems. Inspired by the information relaxation duality, ADRL reformulates the control problem as a min-max optimization between policies and adversarial penalties, enforcing non-anticipativity while preserving optimality. Numerical experiments demonstrate ADRL's superior performance to yield tight dual gaps. Our results highlight the potential of ADRL as a robust computational framework for high-dimensional stochastic control in simulation-based optimization contexts. |
| title | Adversarial Reinforcement Learning: A Duality-Based Approach To Solving Optimal Control Problems |
| topic | Optimization and Control |
| url | https://arxiv.org/abs/2506.00801 |