Abstract
In this paper we address the problem of grounded weights initialization for Recurrent Neural Networks. Specifically, we propose a method, rooted in the field of Random Matrix theory, to perform a fast initialization of recurrent weight matrices that meet specific constraints on their spectral radius. Focusing on the Reservoir Computing (RC) framework, the proposed approach allows us to overcome the typical computational bottleneck related to the eigendecomposition of large matrices, enabling to efficiently design large reservoir networks and hence to address time-series tasks characterized by medium/big datasets. Experimental results show that the proposed method enables an accurate control of the spectral radius of randomly initialized recurrent matrices, providing an initialization approach that is extremely more efficient compared to common RC practice.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Piano-midi.de (www.piano-midi.de); MuseData (www.musedata.org); JSBchorales (chorales by J. S. Bach); Nottingham (ifdo.ca/~seymour/nottingham/nottingham.html).
- 2.
References
Bai, Z., Yin, Y.: Limiting behavior of the norm of products of random matrices and two problems of Geman-Hwang. Probab. Theory Relat. Fields 73(4), 555–569 (1986)
Bordenave, C., Caputo, P., Chafaï, D., Tikhomirov, K.: On the spectral radius of a random matrix. arXiv preprint arXiv:1607.05484 (2016)
Boulanger-Lewandowski, N., Bengio, Y., Vincent, P.: Modeling temporal dependencies in high-dimensional sequences: application to polyphonic music generation and transcription. In: Proceedings of the 29th International Conference on Machine Learning (2012)
Gallicchio, C., Micheli, A.: Architectural and Markovian factors of echo state networks. Neural Netw. 24(5), 440–456 (2011)
Gallicchio, C., Micheli, A., Pedrelli, L.: Deep reservoir computing: a critical experimental analysis. Neurocomputing 268, 87–99 (2017)
Gallicchio, C., Micheli, A., Pedrelli, L.: Design of deep echo state networks. Neural Netw. 108, 33–47 (2018)
Gallicchio, C., Micheli, A., Pedrelli, L.: Comparison between DeepESNs and gated RNNs on multivariate time-series prediction. In: Proceedings of the 27th European Symposium on Artificial Neural Networks (ESANN) (in press)
Jaeger, H., Haas, H.: Harnessing nonlinearity: predicting chaotic systems and saving energy in wireless communication. Science 304(5667), 78–80 (2004)
Lukoševičius, M., Jaeger, H.: Reservoir computing approaches to recurrent neural network training. Comput. Sci. Rev. 3(3), 127–149 (2009)
Pascanu, R., Mikolov, T., Bengio, Y.: On the difficulty of training recurrent neural networks. In: Proceedings of International Conference on Machine Learning (ICML), vol. 28, pp. 1310–1318 (2013)
Sutskever, I., Martens, J., Dahl, G., Hinton, G.: On the importance of initialization and momentum in deep learning. In: International Conference on Machine Learning, pp. 1139–1147 (2013)
Tao, T., Vu, V., Krishnapur, M., et al.: Random matrices: universality of ESDs and the circular law. Ann. Probab. 38(5), 2023–2065 (2010)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Gallicchio, C., Micheli, A., Pedrelli, L. (2020). Fast Spectral Radius Initialization for Recurrent Neural Networks. In: Oneto, L., Navarin, N., Sperduti, A., Anguita, D. (eds) Recent Advances in Big Data and Deep Learning. INNSBDDL 2019. Proceedings of the International Neural Networks Society, vol 1. Springer, Cham. https://doi.org/10.1007/978-3-030-16841-4_39
Download citation
DOI: https://doi.org/10.1007/978-3-030-16841-4_39
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-16840-7
Online ISBN: 978-3-030-16841-4
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)