Model evaluation method based on multi-agent Q learning
A multi-agent and model technology, applied in the field of multi-agent Q-learning models, can solve the problems of not knowing the practicality of theoretical models, inability to correctly evaluate theoretical models, complex control, etc., so as to avoid later impact on use, simple and accurate evaluation The effect of model evaluation
Pending Publication Date: 2021-08-20
WUHAN UNIV OF TECH
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0002] At present, most search and rescue personnel use AUV (Autonomous Underwater Vehicle, autonomous underwater glider) to search and rescue the wrecked sea area. Since the control of the overall queue of large-scale underwater glider is relatively complicated, we propose a multi-agent based , The theoretical model of Q-learning, which performs multi-layer distributed control of the overall queue, but in use, it is not clear which theoretical model is more practical, and it is impossible to make a correct evaluation of the two theoretical models. Therefore, we propose a method based on A Model Evaluation Method for Multi-Agent Q-Learning to Address the Issues Above
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreExamples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0017] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.
[0018] A model evaluation method based on multi-agent Q learning, the specific steps are as follows:
[0019] S1. Establishing a model: establish an initial design model based on the initial data and in accordance with the modeling requirements;
[0020] S2. Model data collection: collect and classify the important data of each part of the initially established model, and part of the data is calculated according to the corresponding formula;
[0021] S3. Model data analysis: conduct theoretical calculations on the collected data, and compare them with the requirements of the modeling data;
[0022] S4. Model function...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a model evaluation method based on multi-agent Q learning. The model evaluation method based on multi-agent Q learning comprises the following specific steps: S1, model establishment, S2, model data acquisition, S3, model data analysis, S4, model function demonstration, S5, model data adjustment, S6, secondary demonstration, and S7, evaluation result obtaining. According to the invention, the model for Q learning can be conveniently evaluated, the practical model can be accurately mastered, the influence on the practical model is avoided, specific parameters of the model can be clearly understood by collecting model data, the use of the model can be intuitively understood by setting model function demonstration, and by setting and adjusting the model data, the maximization of the model function can be realized, and the bearing range of the model can be intuitively understood by setting secondary demonstration.
Description
technical field [0001] The invention relates to the technical field of multi-agent Q-learning models, in particular to a model evaluation method for multi-agent Q-learning. Background technique [0002] At present, most search and rescue personnel use AUV (Autonomous Underwater Vehicle, autonomous underwater glider) to search and rescue the wrecked sea area. Since the control of the overall queue of large-scale underwater glider is relatively complicated, we propose a multi-agent based , The theoretical model of Q-learning, which performs multi-layer distributed control of the overall queue, but in use, it is not clear which theoretical model is more practical, and it is impossible to make a correct evaluation of the two theoretical models. Therefore, we propose a method based on A Model Evaluation Method for Multi-Agent Q-Learning to Address the Issues Above. Contents of the invention [0003] The purpose of the present invention is to provide a model evaluation method b...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F30/27G06N20/00
CPCG06N20/00G06F30/27
Inventor 张磊冯俊尧戎智张竣皓
Owner WUHAN UNIV OF TECH