Improving reinforcement learning with interactive feedback and affordances.
Francisco CruzSven MaggCornelius WeberStefan WermterPublished in: ICDL-EPIROB (2014)
Keyphrases
- reinforcement learning
- user interaction
- state space
- learning algorithm
- user feedback
- markov decision processes
- collaborative learning
- function approximation
- computer graphics
- graphical interface
- temporal difference
- action selection
- model free
- multi agent
- information retrieval
- temporal difference learning
- autonomous learning
- real world
- learning opportunities
- virtual reality
- relevance feedback