Zero-delay rate-distortion optimization for partially observable Gauss-Markov processes.
Takashi TanakaPublished in: CDC (2015)
Keyphrases
- markov processes
- partially observable
- markov chain
- state space
- markov decision processes
- reinforcement learning
- dynamical systems
- decision problems
- rate distortion
- markov process
- stochastic processes
- bit rate
- rate control
- infinite horizon
- video coding
- belief state
- random fields
- inter frame
- non stationary
- planning domains
- stationary distribution
- steady state
- reward function
- search space
- motion estimation
- random walk
- optimal solution