Neural Network Computations with Negative Triggering Thresholds
Recent binary signal detection theoryand neural network assembly memory model’s optimal data-decoding/memory-retrieval algorithm exists simultaneously in functionally equivalent neural network (NN), convolutional, and Hamming distance forms. In present paper this NN algorithm has been specified to provide decoding/retrieval probabilities at both positive and negative neuron triggering thresholds needed, in particular, for ROC curve computations. Examples of intact and damaged NNs are considered, model neuron receptive fields are introduced, a comparison between NN and analytic computations of decoding/retrieval probabilities is also performed.
KeywordsNeural Network Algorithm Bidirectional Associative Memory Trigger Threshold Binarization Rule Neural Network Computation
Unable to display preview. Download preview PDF.
- 4.Gopych, P.M.: Determination of Memory Performance. JINR Rapid Communications 4-99, 61–68 (1999) (in Russian)Google Scholar
- 5.Gopych, P.M.: Sensitivity and Bias within the Binary Signal Detection Theory, BSDT. Int. J. Inf. Theor. & Appl. 11, 318–328 (2004)Google Scholar
- 6.Gopych, P.M.: ROC Curves within the Framework of Neural Network Assembly Memory Model: Some Analytic Results. Int. J. Inf. Theor. & Appl. 10, 189–197 (2003)Google Scholar
- 8.Gopych, P.M.: Identification of Peaks in Line Spectra Using the Algorithm Imitating the Neural Network Operation. Instr. & Exper. Techn. 41, 341–346 (1998)Google Scholar