Corpus ID: 211259324

Deep Reinforcement Learning with Linear Quadratic Regulator Regions

@article{Fernandez2020DeepRL,
title={Deep Reinforcement Learning with Linear Quadratic Regulator Regions},
author={Gabriel Fernandez and Colin Togashi and D. Hong and Lin Yang},
journal={ArXiv},
year={2020},
volume={abs/2002.09820}
}
Practitioners often rely on compute-intensive domain randomization to ensure reinforcement learning policies trained in simulation can robustly transfer to the real world. Due to unmodeled nonlinearities in the real system, however, even such simulated policies can still fail to perform stably enough to acquire experience in real environments. In this paper we propose a novel method that guarantees a stable region of attraction for the output of a policy trained in simulation, even for highly… Expand
1 Citations

References

SHOWING 1-10 OF 33 REFERENCES
Assessing Transferability From Simulation to Reality for Reinforcement Learning
• Computer Science, Medicine
• IEEE Transactions on Pattern Analysis and Machine Intelligence
• 2021
• 11
• PDF
Robust Model-free Reinforcement Learning with Multi-objective Bayesian Optimization
• Computer Science
• 2020 IEEE International Conference on Robotics and Automation (ICRA)
• 2020
• 8
• PDF
Sim-to-Real Transfer of Robotic Control with Dynamics Randomization
• Computer Science, Engineering
• 2018 IEEE International Conference on Robotics and Automation (ICRA)
• 2018
• 435
• PDF
Benchmarking Deep Reinforcement Learning for Continuous Control
• Computer Science, Mathematics
• ICML
• 2016
• 976
• PDF
On the design of LQR kernels for efficient controller learning
• Computer Science, Mathematics
• 2017 IEEE 56th Annual Conference on Decision and Control (CDC)
• 2017
• 18
• PDF
Addressing Function Approximation Error in Actor-Critic Methods
• Computer Science, Mathematics
• ICML
• 2018
• 745
• Highly Influential
• PDF