Mapping Instructions and Visual Observations to Actions with Reinforcement Learning.
Dipendra Kumar MisraJohn LangfordYoav ArtziPublished in: CoRR (2017)
Keyphrases
- reinforcement learning
- perceptual aliasing
- action selection
- partially observable domains
- visual information
- state transitions
- visual features
- sensory inputs
- high level
- action space
- reward function
- function approximation
- model free
- partially observable
- multiagent reinforcement learning
- low level
- state action
- learning process
- learning algorithm
- multi agent
- partial observations
- state and action spaces
- continuous state
- partial observability
- state space
- reinforcement learning algorithms
- temporal difference
- visual perception
- human activities