Personalized deep learning method for fog computing environment

Abstract

The invention discloses a personalized deep learning method for fog computing environment. A cloud node trains a general model through mass training data, the general model obtained by training is distributed to each fog computing node, the computation and storage ability of the fog computing node is used for training a deep leaning model which meets edge side industry requirements; data is collected from intelligent sensing equipment, and real-time reasoning is carried out on the fog computing node to output a result in real time; and errors which appear in reasoning are identified, the modelcan be continuously trained and optimized, the industrial personalized model can be selectively transmitted into the cloud, and meanwhile, the general model of the cloud is received and continuouslyimproved and optimized. Compared with the prior art, the method disclosed by the invention improves real-time business execution efficiency, meanwhile, the personalized deep learning method is storedin the fog computing node to guarantee the safety of the model, in addition, the model can be shared to the cloud according to the permission of the user, and therefore, the model exhibits openness.

Description

technical field [0001] The invention relates to cloud computing, Internet of Things, artificial intelligence and deep learning technologies, in particular to a personalized deep learning method for a fog computing environment. Background technique [0002] In recent years, artificial intelligence technology has developed rapidly, and its commercialization speed has exceeded expectations. Artificial intelligence will bring subversive changes to the entire society, and it has become an important development strategy for various countries in the future. In particular, the algorithm evolution with deep learning as the core, its super evolutionary ability, with the support of big data, through training and construction of a large-scale convolutional neural network similar to the structure of the human brain, can already solve various problems. [0003] Deep learning requires a large amount of data and computing resources for training. Cloud services can meet the requirements to a...

AI Technical Summary

Problems solved by technology

[0004] With the development of deep learning technology, cloud training and learning will produce general data models such as object detection. However, in specific application scenarios, general models cannot meet the individual needs of the industry. Training can be completed in the cloud, and fog computing nodes run through Between the cloud and the device end, it becomes a bridge between the cloud and the device end, and can provide near-end training and reasoning computing services nearby

Images