面向机器学习的隐私保护关键技术研究

Research on Key Technologies of Privacy Protection for Machine Learning

现阶段,随着机器学习技术的快速发展,该技术已经在图像处理、语音识别、智能检测等领域得到了充分的应用。但由于机器学习的过程中会采集大量的数据样本,以保证该模型的应用效果,会导致各种隐私数据随着收集、处理、共享等流程而造成隐私泄露。而另一方面,机器学习模型本身也面临着一定的隐私泄露风险,如隐私攻击。因此,面向机器学习过程中展开隐私保护关键技术的研究,成为机器学习技术发展的重要基础。

At the present stage,with the rapid development of machine learning technology,this technology has been fully applied in image processing,speech recognition,intelligent detection and other fields.However,because a large number of data samples will be collected in the process of machine learning to ensure the application effect of the model,it will lead to the privacy leakage of various privacy data along with the collection,processing,sharing and other processes.On the other hand,machine learning models themselves also face certain privacy leakage risks,such as privacy attacks.Therefore,the research on key technologies of privacy protection in the process of machine learning has become an important foundation for the development of machine learning technology.