Learning to Generate Combinatorial Action Sequences Utilizing the Initial Sensitivity of Deterministic Dynamical Systems

@article{Nishimoto2003LearningTG,
  title={Learning to Generate Combinatorial Action Sequences Utilizing the Initial Sensitivity of Deterministic Dynamical Systems},
  author={Ryunosuke Nishimoto and Jun Tani},
  journal={Neural networks : the official journal of the International Neural Network Society},
  year={2003},
  volume={17 7},
  pages={925-33}
}
This study shows how sensory-action sequences of imitating finite state machines (FSMs) can be learned by utilizing the deterministic dynamics of recurrent neural networks (RNNs). Our experiments indicated that each possible combinatorial sequence can be recalled by specifying its respective initial state value and also that fractal structures appear in this initial state mapping after the learning converges. We also observed that the sequences of mimicking FSMs are encoded utilizing the… CONTINUE READING

Citations

Publications citing this paper.
Showing 1-10 of 23 extracted citations

References

Publications referenced by this paper.
Showing 1-10 of 16 references

Exploring the computational capabilities of recurrent neural networks

  • I. Tsuda
  • distal teacher. Cognitive Science,
  • 2003
1 Excerpt

Effects of repetitive transcranial magnetic stimulation on movementrelated cortical activity in human

  • S. Rossi
  • Cereb. Cortex, 10, 802–808.
  • 2000
2 Excerpts

Similar Papers

Loading similar papers…