论文标题
关于班级选择性，维度和鲁棒性之间的关系
On the relationship between class selectivity, dimensionality, and robustness
论文作者
论文摘要
虽然深度神经网络（DNN）中稀疏和分布式表示之间的相对权衡却进行了充分研究，但对这些权衡如何适用于语义上敏感信息的表示方式却少了。类选择性，单位跨数据类别或维度的响应的可变性是量化语义表示稀疏性的一种方法。鉴于最近的证据表明，类选择性会损害概括，我们试图研究它是否也赋予了鲁棒性（或脆弱性）对输入数据的扰动。我们发现，阶级选择性预测了自然主义腐败的脆弱性。正规化的网络具有较低的班级选择性对损坏更为强大，而具有较高类选择性的网络更容易受到腐败的影响，如使用Tiny ImagenetC和Cifar10c测量的那样。相比之下，我们发现类选择性增加了对基于梯度的对抗攻击的多种类型的鲁棒性。为了检查这种差异，我们研究了由于扰动而导致表示形式变化的维度，发现降低类选择性会增加两种腐败类型的变化的维度，但对对抗性攻击的增加幅度较大。这些结果证明了选择性与鲁棒性之间的因果关系，并为这种关系的机制提供了新的见解。
While the relative trade-offs between sparse and distributed representations in deep neural networks (DNNs) are well-studied, less is known about how these trade-offs apply to representations of semantically-meaningful information. Class selectivity, the variability of a unit's responses across data classes or dimensions, is one way of quantifying the sparsity of semantic representations. Given recent evidence showing that class selectivity can impair generalization, we sought to investigate whether it also confers robustness (or vulnerability) to perturbations of input data. We found that mean class selectivity predicts vulnerability to naturalistic corruptions; networks regularized to have lower levels of class selectivity are more robust to corruption, while networks with higher class selectivity are more vulnerable to corruption, as measured using Tiny ImageNetC and CIFAR10C. In contrast, we found that class selectivity increases robustness to multiple types of gradient-based adversarial attacks. To examine this difference, we studied the dimensionality of the change in the representation due to perturbation, finding that decreasing class selectivity increases the dimensionality of this change for both corruption types, but with a notably larger increase for adversarial attacks. These results demonstrate the causal relationship between selectivity and robustness and provide new insights into the mechanisms of this relationship.
