• Published 2007

Regionalized Policy Representation for Reinforcement Learning in POMDPs

@inproceedings{Liao2007RegionalizedPR,
  title={Regionalized Policy Representation for Reinforcement Learning in POMDPs},
  author={Xuejun Liao and Hui Li and Ronald E. Parr and Lawrence Carin},
  year={2007}
}
Many decision-making problems can be formulated in the framework of a partially observable Markov decision process (POMDP) [5]. The optimality of decisions relies on the accuracy of the POMDP model as well as the policy found for the model. In many applications the model is unknown and learned empirically based on experience, and building a model is just as difficult as finding the associated policy. Since the ultimate goal of decision making is the optimal policy, it is advantageous to learn… CONTINUE READING

Tables from this paper.

Citations

Publications citing this paper.

Multi-task Reinforcement Learning in Partially Observable Stochastic Environments

VIEW 1 EXCERPT
CITES METHODS

References

Publications referenced by this paper.
SHOWING 1-10 OF 10 REFERENCES

Planning and Acting in Partially Observable Stochastic Domains

VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

The State of Mind: Reinforcement Learning with Recurrent Neural Networks

  • B. Bakker
  • PhD thesis,
  • 2004
VIEW 1 EXCERPT

Utile distinction hidden Markov models

VIEW 1 EXCERPT

Infinite-Horizon Policy-Gradient Estimation

VIEW 1 EXCERPT

Reinforcement Learning: An Introduction

VIEW 1 EXCERPT