Lyapunov-Driven Deep Reinforcement Learning for Edge Inference Empowered by Reconfigurable Intelligent Surfaces.
Kyriakos StylianopoulosMattia MerluzziPaolo Di LorenzoGeorge C. AlexandropoulosPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- learning capabilities
- adaptive control
- low cost
- high curvature
- three dimensional
- data driven
- function approximation
- learning algorithm
- inference process
- control theory
- free form
- edge information
- bayesian inference
- probabilistic inference
- optimal policy
- multi agent
- state space
- class of nonlinear systems
- surface reconstruction
- closed loop
- reconfigurable architecture
- asymptotic stability
- hardware implementation
- stability analysis
- edge detection
- reinforcement learning algorithms
- intelligent systems
- temporal difference
- belief nets
- nonlinear systems
- model free
- optimal control
- dynamical systems
- range data
- sufficient conditions
- control law
- function approximators
- transfer learning
- neural network
- functional units
- belief networks