Title :
Invariant Set of Weight of Perceptron Trained by Perceptron Training Algorithm
Author :
Ho, Charlotte Yuk-Fan ; Ling, Bingo Wing-Kuen ; Iu, Herbert Ho-Ching
Author_Institution :
Sch. of Math. Sci., Univ. of London, London, UK
Abstract :
In this paper, an invariant set of the weight of the perceptron trained by the perceptron training algorithm is defined and characterized. The dynamic range of the steady-state values of the weight of the perceptron can be evaluated by finding the dynamic range of the weight of the perceptron inside the largest invariant set. In addition, the necessary and sufficient condition for the forward dynamics of the weight of the perceptron to be injective, as well as the condition for the invariant set of the weight of the perceptron to be attractive, is derived.
Keywords :
learning (artificial intelligence); pattern recognition; perceptrons; set theory; forward dynamics; largest invariant set; perceptron training algorithm; perceptron weight; steady state value; Chaos; Dynamic range; Face recognition; Image recognition; Industrial training; Neural networks; Pattern recognition; Probability; Speech recognition; Vehicle dynamics; Chaos; invariant set; neurodynamics; perceptron training algorithm; symbolic dynamics; Algorithms; Artificial Intelligence; Computer Simulation; Decision Support Techniques; Models, Theoretical; Pattern Recognition, Automated;
Journal_Title :
Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions on
DOI :
10.1109/TSMCB.2010.2042444
