Learned graphical models for probabilistic planning provide a new class of movement primitives.
Elmar A. RückertGerhard NeumannMarc ToussaintWolfgang MaassPublished in: Frontiers Comput. Neurosci. (2012)
Keyphrases
- graphical models
- probabilistic model
- probabilistic planning
- belief propagation
- bayesian networks
- probabilistic graphical models
- chain graphs
- random variables
- probabilistic inference
- conditional random fields
- approximate inference
- structure learning
- heuristic search
- message passing
- influence diagrams
- markov networks
- parameter estimation
- dynamic programming
- hidden markov models