Abstract
Reinforcement learning (RL) algorithms that employ neural networks as function approximators have proven to be powerful tools for solving optimal control problems. However, neural network function approximators suffer from a number of problems like learning becomes difficult when the training data are given sequentially, difficult to determine structural parameters, and usually result in local minima or overfitting. In this paper, a novel on-line sequential learning evolving neural network model design for RL is proposed. We explore the use of minimal resource allocation neural network (mRAN), and develop a mRAN function approximation approach to RL systems. Potential of this approach is demonstrated through a case study. The mean square error accuracy, computational cost, and robustness properties of this scheme are compared with static structure neural networks.
Keywords
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Sutton, R.S., Barto, A.G., Williams, R.J.: Reinforcement learning is direct adaptive optimal control. IEEE Control Syst. Mag. 12(2), 19–22
Sutton, R.S., Barto, A.G.: Reinforcement learning: An introduction. MIT Press, Cambridge
Watkins CJCHLearning with delayed rewards. Ph. D. Thesis, University of Cambridge (1989)
Singh, S., Jaakkola, T., Littman, M., Szpesvari, C.: Convergence results for single step on-policy reinforcement learning algorithms. Machine Learning 38, 287–308 (2000)
Hagen, S.T., Kröse, B.: Neural Q-learning. Neural Comput. & Applic. 12, 81–88 (2003)
Platt, J.: A resource-allocating network for function interpolation. Neural Computation 3, 213–225 (1991)
Kadirkamanathan, V., Niranjan, M.: A function estimation approach to sequential learning with neural networks. Neural Computation 5, 954–975 (1993)
Yingwei, L., Sundararajan, N., Saratchandran, P.: A sequential learning scheme for function approximation using minimal radial basis function (RBF) neural networks. Neural Computation 9, 461–478 (1997)
Yingwei, L., Sundararajan, N., Saratchandran, P.: Performance evaluation of a sequential minimal radial basis function (RBF) neural network learning algorithm. IEEE Trans. on Neural Network 9, 308–318 (1998)
Rojas, I., Pomares, H., Bernier, J.L., Ortega, J., Pino, B., Pelayo, F.J., Prieto, A.: Time series analysis using normalized PG-RBF network with regression weights. Neurocomputing 42, 267–285 (2002)
Salmeron, M., Ortega, J., Puntonet, C.G., Prieto, A., Improved, R.A.N.: sequential prediction using orthogonal techniques. Neurocomputing 41, 153–172 (2001)
Huang, G.B., Saratchandran, P., Sundararajan, N.: An efficient sequential learning algorithm for growing and pruning RBF (GAPRBF) networks. IEEE Transcript on System Man and Cybern. B 34, 2284–2292 (2004)
Huang, G.B., Saratchandran, P., Sundararajan, N.: A generalized growing and pruning RBF (GGAP-RBF) neural network for function approximation. IEEE Transcript on Neural Network 16, 57–67 (2005)
Liang, N.Y., Huang, G.B., Saratchandran, P., Sundararajan, N.: A fast and accurate online sequential learning algorithm for feed forward networks. IEEE Trans. on Neural Network 17, 1411–1423 (2006)
Vamplew, P., Ollington, R.: Global versus local constructive function approximation for on-line reinforcement learning. In: Zhang, S., Jarvis, R.A. (eds.) AI 2005. LNCS (LNAI), vol. 3809, pp. 113–122. Springer, Heidelberg (2005)
Shiraga, N., Ozawa, S., Abe, S.: A reinforcement learning algorithm for neural networks with incremental learning ability. In: Proceeding of the 9th International Conference on Neural Information Processing, vol. 5, pp. 2566–2570 (2002)
Kobayashi, M., Zamani, A., Ozawa, S., Abe, S.: Reducing computations in incremental learning for feed-forward neural network with long-term memory. In: Proc. International Joint Conference on Neural Networks, pp. 1989–1994 (2001)
Shah, H., Gopal, M.: A fuzzy decision tree based robust Markov game controller for robot manipulators. International Journal of Automatic and Control 4(4), 417–439 (2010)
Green, S.J.Z.: Dynamics and trajectory tracking control of a two-link robot manipulator. J. Vibration Control 10(10), 1415–1440 (2004)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Shah, H., Gopal, M. (2014). Neural Networks with Online Sequential Learning Ability for a Reinforcement Learning Algorithm. In: Kumar Kundu, M., Mohapatra, D., Konar, A., Chakraborty, A. (eds) Advanced Computing, Networking and Informatics- Volume 1. Smart Innovation, Systems and Technologies, vol 27. Springer, Cham. https://doi.org/10.1007/978-3-319-07353-8_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-07353-8_11
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-07352-1
Online ISBN: 978-3-319-07353-8
eBook Packages: EngineeringEngineering (R0)