We show that combining human prior knowledge with end-to-end learning can improve the robustness of deep neural networks by introducing a part-based model for object classification. We believe that the richer form of annotation helps guide neural networks to learn more robust features without requiring more samples or larger models. Our model combines a part segmentation model with a tiny classifier and is trained end-to-end to simultaneously segment objects into parts and then classify the segmented object. Empirically, our part-based models achieve both higher accuracy and higher adversarial robustness than a ResNet-50 baseline on all three datasets. For instance, the clean accuracy of our part models is up to 15 percentage points higher than the baseline's, given the same level of robustness. Our experiments indicate that these models also reduce texture bias and yield better robustness against common corruptions and spurious correlations. The code is publicly available at https://github.com/chawins/adv-part-model.
翻译:我们证明,将人类先前的知识与端到端学习相结合,通过引入一个基于部分的物体分类模型,可以提高深神经网络的稳健性。我们认为,更富的注解形式有助于引导神经网络学习更稳健的特征,而不需要更多的样品或更大的模型。我们的模型将一个极小的分类模型结合成一个部分分割模型,经过培训的端到端同时将物体分成几个部分,然后对分解对象进行分类。我们基于部分的模型在所有三个数据集中都实现了比ResNet-50基线更高的准确性和对抗性强。例如,我们部分模型的清洁性比基线的精确度高15个百分点,因为同样强。我们的实验表明,这些模型还减少了对普通腐败和虚伪的关联的文本偏差,并产生更好的稳健性。该代码可在https://github.com/chawins/adv-part-model上公开查阅。