Advanced search
1 file | 299.58 KB

Memory versus non-linearity in reservoirs

Author
Organization
Abstract
Reservoir Computing (RC) is increasingly being used as a conceptually simple yet powerful method for using the temporal processing of recurrent neural networks (RNN). However, because fundamental insight in the exact functionality of the reservoir is as yet still lacking, in practice there is still a lot of manual parameter tweaking or brute-force searching involved in optimizing these systems. In this contribution we aim to enhance the insights into reservoir operation, by experimentally studying the interplay of the two crucial reservoir properties, memory and non-linear mapping. For this, we introduce a novel metric which measures the deviation of the reservoir from a linear regime and use it to define different regions of dynamical behaviour. Next, we study the relationship of two important reservoir parameters, input scaling and spectral radius, on two properties of an artificial task, namely memory and non- linearity.
Keywords
Reservoir Computing, machine learning, neural networks

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 299.58 KB

Citation

Please use this url to cite or link to this publication:

Chicago
Verstraeten, David, Joni Dambre, Xavier Dutoit, and Benjamin Schrauwen. 2010. “Memory Versus Non-linearity in Reservoirs.” In IEEE International Joint Conference on Neural Networks (IJCNN). New York, NY, USA: IEEE.
APA
Verstraeten, D., Dambre, J., Dutoit, X., & Schrauwen, B. (2010). Memory versus non-linearity in reservoirs. IEEE International Joint Conference on Neural Networks (IJCNN). Presented at the 2010 IEEE International joint conference on Neural Networks (IJCNN 2010) ; World congress on Computational Intelligence (WCCI 2010), New York, NY, USA: IEEE.
Vancouver
1.
Verstraeten D, Dambre J, Dutoit X, Schrauwen B. Memory versus non-linearity in reservoirs. IEEE International Joint Conference on Neural Networks (IJCNN). New York, NY, USA: IEEE; 2010.
MLA
Verstraeten, David, Joni Dambre, Xavier Dutoit, et al. “Memory Versus Non-linearity in Reservoirs.” IEEE International Joint Conference on Neural Networks (IJCNN). New York, NY, USA: IEEE, 2010. Print.
@inproceedings{1024070,
  abstract     = {Reservoir	Computing	(RC)	is increasingly being used as a conceptually simple yet powerful method for using the temporal processing of recurrent neural networks (RNN). However, because fundamental insight in the exact functionality of the reservoir is as yet still lacking, in practice there is still a lot of manual parameter tweaking or brute-force searching involved in optimizing these systems. In this contribution we aim to enhance the insights into reservoir operation, by experimentally studying the interplay of the two crucial reservoir properties, memory and non-linear mapping. For this, we introduce a novel metric which measures the deviation of the reservoir from a linear regime and use it to define different regions of dynamical behaviour. Next, we study the relationship of two important reservoir parameters, input scaling and spectral radius, on two properties of an artificial task, namely memory and non- linearity.},
  author       = {Verstraeten, David and Dambre, Joni and Dutoit, Xavier and Schrauwen, Benjamin},
  booktitle    = {IEEE International Joint Conference on Neural Networks (IJCNN)},
  isbn         = {9781424469178},
  issn         = {1098-7576},
  keywords     = {Reservoir Computing,machine learning,neural networks},
  language     = {eng},
  location     = {Barcelona, Spain},
  pages        = {8},
  publisher    = {IEEE},
  title        = {Memory versus non-linearity in reservoirs},
  url          = {http://dx.doi.org/10.1109/IJCNN.2010.5596492},
  year         = {2010},
}

Altmetric
View in Altmetric
Web of Science
Times cited: