Pan-Logical Probabilistic Algorithms Based on Convolutional Neural Networks

Abstract

A brand-new kind of flexible logic system called universal logic aims to address a variety of uncertain problems. In this study, the role of convolutional neural networks in assessing probabilistic pan-logic algorithms is investigated. A generic logic probability algorithm analysis based on a convolutional neural network is suggested due to the unpredictable outputs of the probabilistic algorithm and the difficulty of its analysis. The stochastic gradient descent technique and the error backpropagation algorithm are used to investigate the broad logic probability algorithm (SGD). The experimental data presented in this research show that the BP algorithm of the convolutional neural network has an accuracy rate of 89 percent when analysing the experimental data. As there are more experimental iterations, the error will go down. The SGD method proves that raising the algorithm’s learning rate reduces the loss value of the function. The loss value can be as low as 100%, and the algorithm analysis is closer to the real.