机器学习对抗攻击是一颗“定时炸弹”

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"如果你一直在关注有关人工智能的新闻，你可能已经听说过或看到过经修改的熊猫、海龟和停车标志的图像，这些图像在人们眼中看似平凡，但却能引起人工智能系统的异常行为。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"这些被称为"},{"type":"link","attrs":{"href":"https:\/\/bdtechtalks.com\/2018\/12\/27\/deep-learning-adversarial-attacks-ai-malware\/","title":"","type":null},"content":[{"type":"text","text":"“对抗样本”或“对抗攻击"}]},{"type":"text","text":"”图像及其"},{"type":"link","attrs":{"href":"https:\/\/bdtechtalks.com\/2019\/04\/29\/ai-audio-adversarial-examples\/","title":"","type":null},"content":[{"type":"text","text":"音频"}]},{"type":"text","text":"和文本"},{"type":"link","attrs":{"href":"https:\/\/bdtechtalks.com\/2019\/04\/02\/ai-nlp-paraphrasing-adversarial-attacks\/","title":"","type":null},"content":[{"type":"text","text":"对应物"}]},{"type":"text","text":"，在机器学习领域引起了越来越多的兴趣和关注。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"然而，尽管针对"},{"type":"link","attrs":{"href":"https:\/\/bdtechtalks.com\/2020\/07\/15\/machine-learning-adversarial-examples\/","title":"","type":null},"content":[{"type":"text","text":"对抗机器学习"}]},{"type":"text","text":"的研究不断增加，但数据表明，在解决现实世界应用中的对抗攻击方面进展甚微。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"随着机器学习的迅速普及，科技界必须制定一份保证人工智能系统免遭对抗攻击的路线图。否则的话，对抗机器学习将会是一场灾难。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/59\/7a\/596da7623f1cfc13bbab6c80906b6c7a.jpg","alt":null,"title":"","style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":"","fromPaste":false,"pastePass":false}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"人工智能研究人员发现，通过在停车标志上贴上黑白小贴纸，就能使它们对计算机视觉算法不可见（来源：arxiv.org）。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"是什么让对抗攻击有所不同"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"每种软件都有其独特的安全漏洞，并随着软件发展的新趋势而产生新的威胁。比如， SQL 注入攻击开始流行，因为带数据库后台的 Web 应用开始取代静态网站。由于大量采用浏览器端脚本语言，导致了跨站脚本攻击。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"缓冲区溢出攻击利用 C 等编程语言处理内存分配的方式，覆盖关键变量，在目标计算机上执行恶意代码。反序列化攻击利用了诸如 Java 和 Python 等编程语言在应用程序和进程之间传输信息的方式的缺陷。"}]}]}