Reinforcement Learning With Constrained Uncertain Reward Function Through Particle Filtering.
Oguzhan DogruRanjith ChiplunkarBiao HuangPublished in: IEEE Trans. Ind. Electron. (2022)
Keyphrases
- particle filtering
- reward function
- reinforcement learning
- particle filter
- reinforcement learning algorithms
- state space
- visual tracking
- markov decision processes
- inverse reinforcement learning
- optimal policy
- partially observable
- human motion
- markov decision process
- appearance model
- mean shift
- bayesian filtering
- transition model
- object tracking
- function approximation
- multiple agents
- kalman filter
- generative model
- initially unknown
- learning algorithm
- computer vision
- state variables
- density function
- action selection
- monte carlo
- markov chain
- hidden markov models
- reward signal