Simultaneous estimation of rewards and dynamics from noisy expert demonstrations.
Michael HermanTobias GindeleJörg WagnerFelix SchmittWolfram BurgardPublished in: ESANN (2016)
Keyphrases
- reinforcement learning
- estimation accuracy
- estimation algorithm
- dynamic model
- machine learning
- human experts
- noise free
- robust estimation
- noisy data
- dynamical systems
- missing data
- domain knowledge
- information retrieval
- markov decision processes
- expert knowledge
- estimation error
- monte carlo simulation
- high dimensional
- parametric models
- case study
- learning algorithm