Exploring parameter space in reinforcement learning

@article{Rckstie2010ExploringPS,
  title={Exploring parameter space in reinforcement learning},
  author={Thomas R{\"u}ckstie\ss and Frank Sehnke and Tom Schaul and Daan Wierstra and Yi Sun and J{\"u}rgen Schmidhuber},
  journal={Paladyn},
  year={2010},
  volume={1},
  pages={14-24}
}
This paper discusses parameter-based exploration methods for reinforcement learning. Parameter-based methods perturb parameters of a general function approximator directly, rather than adding noise to the resulting actions. Parameter-based exploration unifies reinforcement learning and black-box optimization, and has several advantages over action perturbation. We review two recent parameter-exploring algorithms: Natural Evolution Strategies and Policy Gradients with Parameter-Based Exploration… CONTINUE READING

Citations

Publications citing this paper.
Showing 1-10 of 44 extracted citations

References

Publications referenced by this paper.
Showing 1-10 of 35 references

Similar Papers

Loading similar papers…