Exploration vs Exploitation with Partially Observable Gaussian Autoregressive Arms.
Julia KuhnMichel MandjesYoni NazarathyPublished in: EAI Endorsed Trans. Self Adapt. Syst. (2015)
Keyphrases
- autoregressive
- partially observable
- state space
- markov decision processes
- decision problems
- dynamical systems
- partial observability
- reinforcement learning
- non stationary
- partial observations
- infinite horizon
- gaussian markov random field
- random fields
- markov decision problems
- belief state
- maximum likelihood
- reward function
- planning domains
- generative model
- linear programming
- sar images
- probabilistic model
- model selection
- dynamic programming
- partially observable markov decision processes
- image segmentation
- orders of magnitude