论文标题
通过隐私/费用理论的强大机器学习
Robust Machine Learning via Privacy/Rate-Distortion Theory
论文作者
论文摘要
出现了强大的机器学习配方，以解决深层神经网络对对抗性例子的普遍脆弱性。我们的工作吸引了最佳的鲁棒学习与隐私 - 实用性权衡问题之间的联系，这是对利率降低问题的概括。可以通过解决最大条件熵问题的解决方案找到强大分类器和对抗扰动之间游戏的鞍点。这种信息理论的观点阐明了稳健性和清洁数据性能之间的基本权衡，这最终源于基础数据分布和扰动约束的几何结构。
Robust machine learning formulations have emerged to address the prevalent vulnerability of deep neural networks to adversarial examples. Our work draws the connection between optimal robust learning and the privacy-utility tradeoff problem, which is a generalization of the rate-distortion problem. The saddle point of the game between a robust classifier and an adversarial perturbation can be found via the solution of a maximum conditional entropy problem. This information-theoretic perspective sheds light on the fundamental tradeoff between robustness and clean data performance, which ultimately arises from the geometric structure of the underlying data distribution and perturbation constraints.
