Adversarial support vector machine

Abstract

The invention provides a generative adversarial support vector machine based on edge computing collaboration and a text image vector processing system based on a generative adversarial network model.The text image vector processing system comprises a text description input module and a collaborative edge computing terminal. The collaborative edge computing terminal generates a long text description based on a plurality of text keywords provided by the text description input module and based on different deep learning models configured for the collaborative edge computing terminal, inputs thelong text description into a classification model containing a support vector machine for classification, outputs image vectors corresponding to different classifications, and returns the image vectorto the plurality of collaborative edge computing terminals, wherein the edge computing terminals generate scene descriptions of different scene categories corresponding to a plurality of different categories of image vectors; and the image vector and the scene description are taken as input of the adversarial network model to obtain multi-object image output. According to the generative adversarial support vector machine based on edge computing collaboration, the high-resolution image containing multiple objects can be generated quickly and accurately.

Description

technical field [0001] The invention belongs to the field of deep learning, and in particular relates to a generation confrontation support vector machine based on edge computing collaboration and a text image vector processing system based on a generation confrontation network model. Background technique [0002] Adversarial machine learning is an intersection of machine learning and computer security. Adversarial machine learning aims to provide security for machine learning techniques in hostile environments. Since machine learning technology generally studies the same or relatively stable data distribution, when it is deployed in reality, due to the existence of malicious users, this assumption may not necessarily hold true. For example, researchers have found that some well-designed adversarial examples can make machine learning models fail to output correct results. [0003] Support Vector Machine (SVM) is a kind of generalized linear classifier (generalized linear c...

AI Technical Summary

Problems solved by technology

[0008] However, for other problems, such as the problem of generating multi-object images based on text, no similar technical solutions have been seen

Images