Active inference and agency: optimal control without cost functions.
Karl J. FristonSpyridon SamothrakisP. Read MontaguePublished in: Biol. Cybern. (2012)
Keyphrases
- optimal control
- cost function
- control problems
- dynamic programming
- feedback control
- control strategy
- reinforcement learning
- search space reduction
- risk sensitive
- class of nonlinear systems
- optimal control problems
- brownian motion
- infinite horizon
- lyapunov function
- control law
- dynamic bayesian networks
- bayesian networks
- machine learning
- multistage
- search space