Regularized Recurrent Neural Networks for Data Efficient Dual-Task Learning
We introduce a regularization technique to improve system identification for dual-task learning with recurrent neural networks. In particular, the method is introduced using the Factored Tensor Recurrent Neural Networks first presented in . Our goal is to identify a dynamical system with few available observations by augmenting them with data from a sufficiently observed similar system. In our previous work, we discovered that the model accuracy degrades whenever little data of the system of interest is available. The presented regularization term in this work allows to significantly reduce the model error thereby improving the exploitation of knowledge of the well observed system. This scenario is crucial in many real world applications, where data efficiency plays an important role. We motivate the problem setting and our regularized dual-task learning approach by industrial use cases, e.g. gas or wind turbine modeling for optimization and monitoring. Then, we formalize the problem and describe our regularization term by which the learning objective of the Factored Tensor Recurrent Neural Network is extended. Finally, we demonstrate its effectiveness on the cart-pole and mountain car benchmarks.
Keywordsmulti-task learning recurrent neural networks factored tensor recurrent neural networks system identification dynamical systems regularization
Unable to display preview. Download preview PDF.
- 1.Spieckermann, S., Düll, S., Udluft, S., Hentschel, A., Runkler, T.: Exploiting similarity in system identification tasks with recurrent neural networks. In: European Symposium on Artificial Neural Networks, ESANN, pp. 473–478 (2014)Google Scholar
- 2.Schäfer, A.M., Udluft, S., Zimmermann, H.G.: The recurrent control neural network. In: European Symposium on Artificial Neural Networks, ESANN, pp. 319–324 (2007)Google Scholar
- 4.Schäfer, A.M., Schneegass, D., Sterzing, V., Udluft, S.: A neural reinforcement learning approach to gas turbine control. In: Proceedings of the International Joint Conference on Neural Networks, pp. 1691–1696 (August 2007)Google Scholar
- 5.Sutskever, I., Martens, J., Hinton, G.E.: Generating text with recurrent neural networks. In: Proceedings of the 28th International Conference on Machine Learning, pp. 1017–1024 (2011)Google Scholar
- 6.Taylor, G.W., Hinton, G.E.: Factored conditional restricted boltzmann machines for modeling motion style. In: Proceedings of the 26th International Conference on Machine Learning, pp. 1025–1032 (2009)Google Scholar
- 7.Florian, R.V.: Correct equations for the dynamics of the cart-pole system. Center for Cognitive and Neural Studies (Coneural), Romania (2007)Google Scholar
- 8.Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction, vol. 1. Cambridge University Press (1998)Google Scholar
- 9.Bergstra, J., Breuleux, O., Bastien, F., Lamblin, P., Pascanu, R., Desjardins, G., Turian, J., Warde-Farley, D., Bengio, Y.: Theano: a CPU and GPU math expression compiler. In: Proceedings of the Python for Scientific Computing Conference (SciPy). Oral Presentation (2010)Google Scholar
- 10.Martens, J., Sutskever, I.: Learning recurrent neural networks with Hessian-Free Optimization. In: Proceedings of the 28th International Conference on Machine Learning, pp. 1033–1040 (2011)Google Scholar
- 11.Martens, J., Sutskever, I.: Training deep and recurrent networks with Hessian-Free Optimization. In: Montavon, G., Orr, G.B., Müller, K.-R. (eds.) NN: Tricks of the Trade, 2nd edn. LNCS, vol. 7700, pp. 479–535. Springer, Heidelberg (2012)Google Scholar
- 12.Düll, S., Hans, A., Udluft, S.: The markov decision process extraction network. In: European Symposium on Artificial Neural Networks, ESANN (2010)Google Scholar