Deep neural networks are currently the most widespread and successful technology in artificial intelligence. However, these systems exhibit bewildering new vulnerabilities: most notably a susceptibility to adversarial examples. Here, I review recent empirical research on adversarial examples that suggests that deep neural networks may be detecting in them features that are predictively useful, though inscrutable to humans. To understand the implications of this research, we should contend with some older philosophical puzzles about scientific reasoning, helping us to determine whether these features are reliable targets of scientific investigation or just the distinctive processing artefacts of deep neural networks.

深度神经网络是当前人工智能中最广泛，最成功的技术。但是，这些系统表现出令人困惑的新漏洞：最明显的是对抗性示例的易感性。在这里，我回顾了有关对抗性示例的最新实证研究，这些研究表明，深层神经网络可能会在其中检测到可预测的有用功能，尽管这些功能对人类来说是不可理解的。为了理解这项研究的意义，我们应该应对一些有关科学推理的较旧的哲学难题，以帮助我们确定这些特征是科学研究的可靠目标还是仅仅是深层神经网络的独特加工伪像。