Physical deep learning based on optimal control of dynamical systems.
Genki FuruhataTomoaki NiiyamaSatoshi SunadaPublished in: CoRR (2020)
Keyphrases
- dynamical systems
- optimal control
- deep learning
- linear quadratic
- dynamic programming
- control problems
- differential equations
- unsupervised learning
- control theory
- linear systems
- machine learning
- nonlinear dynamical systems
- control strategy
- reinforcement learning
- infinite horizon
- control law
- state space
- optimal control problems
- weakly supervised
- mental models
- markov decision processes
- predictive state representations
- hamilton jacobi bellman