Evolino: Hybrid Neuroevolution/Optimal Linear Search for Sequence Learning

@inproceedings{Schmidhuber2005EvolinoHN,
  title={Evolino: Hybrid Neuroevolution/Optimal Linear Search for Sequence Learning},
  author={J{\"u}rgen Schmidhuber and Daan Wierstra and Faustino J. Gomez},
  booktitle={IJCAI},
  year={2005}
}
Current Neural Network learning algorithms are limited in their ability to model non-linear dynamical systems. Most supervised gradient-based recurrent neural networks (RNNs) suffer from a vanishing error signal that prevents learning from inputs far in the past. Those that do not, still have problems when there are numerous local minima. We introduce a general framework for sequence learning, EVOlution of recurrent systems with LINear outputs (Evolino). Evolino uses evolution to discover good… CONTINUE READING
Highly Cited
This paper has 66 citations. REVIEW CITATIONS

Citations

Publications citing this paper.
Showing 1-10 of 31 extracted citations

Decision Making in Multiagent Systems: A Survey

IEEE Transactions on Cognitive and Developmental Systems • 2018
View 4 Excerpts
Highly Influenced

New Millennium AI and the Convergence of History

Challenges for Computational Intelligence • 2007
View 6 Excerpts
Method Support
Highly Influenced

66 Citations

01020'07'10'13'16'19
Citations per Year
Semantic Scholar estimates that this publication has 66 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
Showing 1-10 of 14 references

Modeling non-linear dynamical systems with evolino

Wierstraet al, 2005 Daan Wierstra, Juergen Schmidhuber, Faustino Gomez
2005

A generalized inverse for matrices

R. Penrose
Proceedings of the Cambridge Philosophy Society • 1996

Springer

V. Vapnik.The Nature of Statistical Learning Theory
New York, • 1995
View 2 Excerpts