关于人工智能的黑色思考-当人工智能遇到安全.pdf

1、关于人工智能的黑色思考-当人工智能遇到安全美国乔治亚大学教授乔治亚网络安全与隐私学院院长安全智能从安全的角度审视机器学习智能安全安全对抗中的人工智能和挑战思考一：AI与机器学习的成功Thought 1:The Success of AI人工智能深度学习的成功应用利用机器学习刷帖刷评价“Leverage deep learning language models(Recurrent Neural Networks or RNNs)to automate the generation of fake online reviews for products and serviceshttps:/ 2

2、:The Opposite of AIArtificial Intelligence（人工智能）Natural Stupidity*（天然傻缺）The Opposite of Artificial Intelligence*The term was initially used by Drew McDermott in his 1976 paper“Artificial Intelligence meets Natural Stupidity”,http:/dl.acm.org/citation.cfm?id=1045340Image Recognition:Flagship of AI Ap

3、plicationshttps:/www.cs.toronto.edu/ranzato/publications/taigman_cvpr14.pdf深度学习对手写识别的应用(MNIST dataset）http:/ dataset）利用模型分类0输入http:/ dataset）Error Rate:0.4%http:/ IntelligenceNatural Stupidity关于识别结果的问题可能的结果：0 1,2,9 App hangs App gets owned Not sure 利用模型分类输入FileAPP思考二:AI 敌不过 Natural Stupidity最近发现的相关漏

4、洞（by 360 Team）OpenCVCVE-2017-12597CVE-2017-12598CVE-2017-12599CVE-2017-12600CVE-2017-12601CVE-2017-12602CVE-2017-12603CVE-2017-12604CVE-2017-12605CVE-2017-12606NumpyCVE-2017-12852OpenEXRCVE-2017-12596LibjasperCVE-2017-9782 人工智能应用实现细节 Complexity and Dependency Stupidity lead to Software Vulnerability

5、 详细漏洞实例“基于深度学习的智能系统中的安全漏洞及影响”，肖奇学The Devil is in the Detail思考三：从安全的角度考虑准确率Thought 3:The Probability GameThe Probability GameMNIST(NYU/Google Labs)Accuracy:99.6%DeepFace(Facebook)Accuracy:97.25%The Probability Game从人工智能的角度看？MNIST(NYU/Google Labs)Accuracy:99.6%DeepFace(Facebook)Accuracy:97.25%Great Ac

6、hievementHuman 97.5%The Probability Game从安全的角度看？MNIST(NYU/Google Labs)Accuracy:99.6%DeepFace(Facebook)Accuracy:97.25%Error rate 0.4%100%Adversarial LearningAdversarial Learning系统化稳定生成对抗样本 白盒/灰盒对抗手段 对抗神经元网络与对抗样本生成 许伟林，“对抗式机器学习中的攻击与防御”黑盒对抗手段利用漏洞挖掘的方法自动生成对抗样本 肖奇学,“基于深度学习的智能系统中的安全漏洞及