부산대학교 도서관

The growing abundance of high-quality data sets, combined with substantial technical developments, has advanced machine learning into a major tool that is employed in a broad array of applications, from cybersecurity to medical diagnosis. The potential attacks against machine learning come in many forms, but the end goal of an adversary is to cause the machine learning model to behave in a manner contrary to the developer's intention. Making Machine Learning Learn the Wrong Thing Most machine learning algorithms require large amounts of data for training purposes. Making Machine Learning Do the Wrong Thing While data poisoning attacks occur during training time, it is possible to attack machine learning models even after they have been developed. [Extracted from the article]