Deep Q-learning: a robust control approach
Varga, Balazs, Kulcsar, Balazs, Chehreghani, Morteza Haghir
–arXiv.org Artificial Intelligence
In this paper, we place deep Q-learning into a control-oriented perspective and study its learning dynamics with well-established techniques from robust control. We formulate an uncertain linear time-invariant model by means of the neural tangent kernel to describe learning. We show the instability of learning and analyze the agent's behavior in frequency-domain. Then, we ensure convergence via robust controllers acting as dynamical rewards in the loss function. We synthesize three controllers: state-feedback gain scheduling H2, dynamic Hinf, and constant gain Hinf controllers. Setting up the learning agent with a control-oriented tuning methodology is more transparent and has well-established literature compared to the heuristics in reinforcement learning. In addition, our approach does not use a target network and randomized replay memory. The role of the target network is overtaken by the control input, which also exploits the temporal dependency of samples (opposed to a randomized memory buffer). Numerical simulations in different OpenAI Gym environments suggest that the Hinf controlled learning performs slightly better than Double deep Q-learning.
arXiv.org Artificial Intelligence
Nov-7-2022
- Country:
- North America > United States
- New York (0.04)
- Maryland > Baltimore (0.04)
- Colorado > Denver County
- Denver (0.04)
- California > Alameda County
- Berkeley (0.04)
- Europe > Sweden
- Vaestra Goetaland > Gothenburg (0.04)
- North America > United States
- Genre:
- Research Report (1.00)
- Technology: