Turnpike in optimal control of PDEs, ResNets, and beyond.
Borjan GeshkovskiEnrique ZuazuaPublished in: Acta Numer. (2022)
Keyphrases
- optimal control
- partial differential equations
- dynamic programming
- control problems
- feedback control
- solving partial differential equations
- control strategy
- level set
- image denoising
- class of nonlinear systems
- optimal control problems
- risk sensitive
- reinforcement learning
- infinite horizon
- numerical solution
- image enhancement
- control law
- image processing
- real time
- brownian motion
- neural network
- stochastic control
- lyapunov function
- denoising
- state space
- data mining
- policy gradient
- machine learning
- learning algorithm
- linear quadratic
- multiscale