Learning of Non-Parametric Control Policies with High-Dimensional State Features

@inproceedings{Hoof2015LearningON,
  title={Learning of Non-Parametric Control Policies with High-Dimensional State Features},
  author={Herke van Hoof and Jan Peters and Gerhard Neumann},
  booktitle={AISTATS},
  year={2015}
}
Learning complex control policies from highdimensional sensory input is a challenge for reinforcement learning algorithms. Kernel methods that approximate values functions or transition models can address this problem. Yet, many current approaches rely on instable greedy maximization. In this paper, we develop a policy search algorithm that integrates robust policy updates and kernel embeddings. Our method can learn nonparametric control policies for infinite horizon continuous MDPs with high… CONTINUE READING
Highly Cited
This paper has 33 citations. REVIEW CITATIONS

Citations

Publications citing this paper.
Showing 1-10 of 24 extracted citations

References

Publications referenced by this paper.
Showing 1-10 of 18 references

Similar Papers

Loading similar papers…