Safe Use of Neural Networks
This addresses the reliability of neural networks in communication systems, but it is incremental as it builds on existing error-detection methods with specific adaptations.
The paper tackles the problem of internal numerical errors in neural networks used in communication systems, which can drastically affect decision results, by employing number-based codes to detect arithmetic errors in processing steps, with mathematical program simulations showing the techniques are effective and efficient.
Neural networks in modern communication systems can be susceptible to internal numerical errors that can drastically effect decision results. Such structures are composed of many sections each of which generally contain weighting operations and activation function evaluations. The safe use comes from methods employing number based codes that can detect arithmetic errors in the network's processing steps. Each set of operations generates parity values dictated by a code in two ways. One set of parities is obtained from a section's outputs while a second comparable set is developed directly from the original inputs. The parity values protecting the activation functions involve a Taylor series approximation to the activation functions. We focus on using long numerically based convolutional codes because of the large size of data sets. The codes are based on Discrete Fourier Transform kernels and there are many design options available. Mathematical program simulations show our error-detecting techniques are effective and efficient.