Q-learning based double-layer Bayes network interference algorithm

Abstract

The invention provides a Q-learning based double-layer Bayes network interference algorithm which comprises the steps of 1) initializing a Q-learning probability table of nodes; 2) updating the condition selection probabilities in the horizontal and vertical axes respectively; and 3) determining value combinations in the horizontal axis and nodes in the vertical axis, and deleting redundant value combinations and nodes. According to the Q-learning based double-layer Bayes network interference algorithm, the probability dependence relationship between the double-layer network parameters is molded, subsequent network state is inferred and analyzed according to known network state, uncertain information of obtained network nodes in the inference process is learned and determined via Q-learning and further classified to obtain the reliability values of probability, an obtained double-layer Bayes network model is simplified by only reserving information most useful for inference, and the model is thus more helpful for realization and accurate inference.

Description

technical field [0001] The invention relates to a reasoning algorithm, in particular to a double-layer Bayesian network reasoning algorithm based on a reinforcement learning algorithm. Background technique [0002] Cognition of the network is to adjust the corresponding configuration inside the network to adapt to changes in the external environment by perceiving the external environment and through self-understanding and learning. The cognitive process is a process of continuously learning and accumulating relevant experience in the process of dynamic self-adaptation, and using this as a basis to make relevant adjustments, judgments, and reconfigurations to the network. The adaptive dynamic adjustment process occurs before the problem occurs, not after, so the performance improvement of the network focuses on the end-to-end Quality of Service (QoS) performance of the entire network. Due to the above characteristics, traditional network cognition can provide users with bett...

AI Technical Summary

Problems solved by technology

[0011] (1) Most of these studies are aimed at a certain local and specific control method, and it is difficult to upgrade to a method and mechanism with global significance;

[0012] (2) The existing research results lack a global assessment of the network situation, and lack of understanding of the cognitive ability, knowledge level and other personality characteristics of the network level (learners);

[0013] (3) It cannot fully meet the different needs of users, and cannot provide learners with personalized reconfiguration data support and guidance

Images