Reinforcement learning for mixed-integer problems based on MPC
Reinforcement learning for mixed-integer problems based on MPC
Model Predictive Control has been recently proposed as policy approximation for Reinforcement Learning, offering a path towards safe and explainable Reinforcement Learning. This approach has been investigated for Q-learning and actor-critic methods, both in the context of nominal Economic MPC and Robust (N)MPC, showing very promising results. In that context, actor-critic methods seem to be the most reliable approach. Many applications include a mixture of continuous and integer inputs, for which the classical actor-critic methods need to be adapted. In this paper, we present a policy approximation based on mixed-integer MPC schemes, and propose a computationally inexpensive technique to generate exploration in the mixed-integer input space that ensures a satisfaction of the constraints. We then propose a simple compatible advantage function approximation for the proposed policy, that allows one to build the gradient of the mixed-integer MPC-based policy.
Dokumentinformationen
Reinforcement learning for mixed-integer problems based on MPC
2020-01-01
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
| DDC: | 629 |
Ähnliche Titel
Learning contact-aware robot controllers from mixed integer optimization
| BASE | 2019
A method for mixed integer programming problems by particle swarm optimization
| Online Contents | 2006
Hierarchical Reinforcement Learning Under Mixed Observability
| Springer Verlag | 2022