论文信息 - Neural networks-competence and performance

Neural networks-competence and performance

This paper investigates the competence and performance of neural computers. The discrete neural computing methods can neither break the unsolvability barrier nor can ameliorate the complexity of resource (time and space) consumption in computation. Also the introduction of randomization or probabilities along with neural computing cannot improve this situation. Very little is known about analog neural networks as they have not been axiomatized and studied; if one axiomatizes and studies them, one may derive similar conclusions concerning their competence and performance.

E. V. Krishnamurthy

[1] James L. Johnson. A Neural Network Approach to the 3-Satisfiability Problem , 1989, J. Parallel Distributed Comput..

[2] Vikram Krishnamurthy,et al. An ANN Model Perceptron Algorithm Using Generalized Matrix Inversion , 1994, Parallel Comput..

[3] George Cybenko,et al. Complexity Theory of Neural Networks and Classification Problems , 1990, EURASIP Workshop.

[4] Georg Schwarz,et al. Connectionism, Processing, Memory , 1992 .

[5] D. Gorse,et al. AN ANALYSIS OF NOISY RAM AND NEURAL NETS , 1989 .

[6] E. V. Krishnamurthy. Unsolvability, complexity, and neural networks , 1994, Neural Networks.

[7] Igor Aleksander,et al. A probabilistic logic neuron network for associative learning , 1989 .

[8] Kazuyuki Aihara,et al. Pulse propagation networks: A neural network model that uses temporal coding by action potentials , 1993, Neural Networks.

[9] John J. Hopfield,et al. Neurons, Dynamics and Computation , 1994 .

[10] Max H. Garzon,et al. Cellular automata and discrete neural networks , 1990 .