Abstract
A number of types of neural network have been shown to be useful for a wide range of tasks, and can be “trained” in a large number of ways. This paper considers how it might be possible to train and run neural networks to respond in different ways under different prevailing circumstances, achieving smooth transitions between multiple learned behaviours in a single network. This type of behaviour has been shown to be useful in a range of applications, such as maintenance of homeostasis. We introduce a novel technique for training multilayer perceptrons which improves on the transitional behaviour of many existing methods, and permits explicit training of multiple behaviours in a single network using gradient descent.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Beer, R.D., Gallagher, J.C.: Evolving dynamical neural networks for adaptive behavior. Adapt. Behav. 1(1), 91–122 (1992)
Bishop, C.M.: Neural Networks for Pattern Recognition. Oxford University Press, New York (1995)
Husbands, P., McHale, G.: Quadrupedal locomotion: GasNets, CTRNNs and Hybrid CTRNN/PNNs compared. In: Proceedings of the 9th International Conference on the Simulation and Synthesis of Living Systems (ALIFE IX), pp. 106–112 (2004). http://sro.sussex.ac.uk/16037/
Husbands, P., Smith, T., Shea, M.O., Jakobi, N., Anderson, J., Philippides, A.: Brains, gases and robots. In: ICANN 1998: Proceedings of the 8th International Conference on Artificial Neural Networks, Skövde, Sweden, 2–4 September 1998 (Perspectives in Neural Computing), pp. 51–63 (1998)
Kaczmarek, L.K., Levitan, I.B.: Neuromodulation: the biochemical control of neuronal excitability. Oxford University Press, New York (1987)
Lecun, Y., Cortes, C.: The MNIST database of handwritten digits. http://yann.lecun.com/exdb/mnist/
Magg, S., Philippides, A.: GasNets and CTRNNs – a comparison in terms of evolvability. In: Nolfi, S., et al. (eds.) SAB 2006. LNCS (LNAI), vol. 4095, pp. 461–472. Springer, Heidelberg (2006)
Neal, M., Timmis, J.: Timidity: a useful emotional mechanism for robot control? Informatica (Slovenia) 27(2), 197–204 (2003)
Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning representations by back-propagating errors. Nature 323(6088), 533–536 (1986)
Sauze, C., Neal, M.: Artificial endocrine controller for power management in robotic systems. IEEE Trans. Neural Netw. Learn. Syst. 24(12), 1973–1985 (2013). doi:10.1109/TNNLS.2013.2271094
Smith, T., Husbands, P., Philippides, A., O’Shea, M.: Neuronal plasticity and temporal adaptivity: GasNet robot control networks. Adapt. Behav. 10(3–4), 161–183 (2002)
Vargas, P.A., Di Paolo, E.A., Husbands, P.: A study of GasNet spatial embedding in a delayed-response task. In: Artificial Life XI, Proceedings of the Eleventh International Conference on the Simulation and Synthesis of Living Systems, pp. 640–647 (2008)
Werbos, P.: Beyond regression: new tools for prediction and analysis in the behavioral sciences. Ph.D. thesis, Harvard (1974)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Finnis, J.C., Neal, M. (2016). UESMANN: A Feed-Forward Network Capable of Learning Multiple Functions. In: Tuci, E., Giagkos, A., Wilson, M., Hallam, J. (eds) From Animals to Animats 14. SAB 2016. Lecture Notes in Computer Science(), vol 9825. Springer, Cham. https://doi.org/10.1007/978-3-319-43488-9_10
Download citation
DOI: https://doi.org/10.1007/978-3-319-43488-9_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-43487-2
Online ISBN: 978-3-319-43488-9
eBook Packages: Computer ScienceComputer Science (R0)