Applying Different Learning Rules in Neuro-Symbolic Integration

Abstract:

Pseudo inverse learning rule and new activation unction performance will be evaluated and compared with the primitive learning rule, Hebb rule. Comparisons are made between these three rules to see which rule is better or outperformed other rules in the aspects of computation time, memory and complexity. From the computer simulation that has been carried out, the new activation function performs better than the other two learning methods.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 716-720
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.716
• Online since: January 2012
• Authors: Saratha Sathasivam
• Keywords: Capacity, Hebb Rule, New Activation Function, Pseudo Inverse

References

[1] Kaslik, E. & Balint, S, Configuration of steady states for Hopfield-type neural networks, in Applied Mathematics and Computation, 182(1), 2006, pp.179-186.

DOI: 10.1016/j.amc.2006.04.054

[2] Wan Abdullah, W.A.T., Logic Programming on a Neural Network, in. Int .J. Intelligent Sys, 7, 1992, pp.513-519.

[3] Saratha Sathasivam. 2010. Usage of New Activation Function in Neuro-Symbolic Integration. In 4th Asian Physics Sympossium, Bali Indonesia.

[4] Saratha Sathasivam. 2010. Neuro symbolic integration using pseudo inverse rule. In Annual International Conference on Advance Topics in Artificial Intelligence. Phuket, Thailand. (ISBN: 978-981-08-7654-8), pp A-69-A-73.

[5] Hopfield, J.J., Neural Networks and Physical Systems with Emergent Collective Computational Abilities., in Proceedings. Natl. Acad. Sci. USA. , 79(8), 1982, pp.2554-2558.

DOI: 10.1073/pnas.79.8.2554

[6] Browne, A & Sun. R., Connectionist inference models, in Neural Networks, 14(10), 2001, p.1331–1355.

DOI: 10.1016/s0893-6080(01)00109-5

[7] Saratha Sathasivam. 2009. Enhancing Logic Programming Performance in Recurrent Hopfield Network, European Journal of Scientific Research, 37(1), pp.1-7.

[8] Saratha Sathasivam. 2009. Logical Content in the Recurrent Hopfield Network without Higher Order Connections, European Journal of Scientific Research, 37(3), pp.361-367.

[9] Saratha Sathasivam. 2009. Learning Rule Performance Comparison in Hopfield Network, American Journal of Scientific Research, Issue 6, pp.15-22.

[10] Saratha Sathsivam & Wan Abdullah, W.A.T. 2010. The Satisfiability Aspect of Logic on Little Hopfield Network, American Journal of Scientific Research, Issue 7, pp.90-105.

[11] Saratha Sathasivam. 2010. Upgrading Logic Programming in Hopfield Network, Sains Malaysiana, 39(1), pp.115-118.

[12] Zeng, X. & Martinez, R. A new activation function in the Hopfield Network for Solving Optimization Problems. In Fourth International Conference on Artificial Neural Networks and Genetic Algorithms, 1999, pp.1-5.

DOI: 10.1007/978-3-7091-6384-9_13