ConvNets and ImageNet Beyond Accuracy: Understanding Mistakes and Uncovering Biases
This study addresses the reliability and bias issues of ConvNets for end-users, offering insights into improving trust in these models.
This paper investigates the reliability of ConvNets on ImageNet, finding that their accuracy and robustness are underestimated. It also demonstrates that explanations can reduce the impact of misclassified adversarial examples and introduces a tool to uncover model biases.
ConvNets and Imagenet have driven the recent success of deep learning for image classification. However, the marked slowdown in performance improvement combined with the lack of robustness of neural networks to adversarial examples and their tendency to exhibit undesirable biases question the reliability of these methods. This work investigates these questions from the perspective of the end-user by using human subject studies and explanations. The contribution of this study is threefold. We first experimentally demonstrate that the accuracy and robustness of ConvNets measured on Imagenet are vastly underestimated. Next, we show that explanations can mitigate the impact of misclassified adversarial examples from the perspective of the end-user. We finally introduce a novel tool for uncovering the undesirable biases learned by a model. These contributions also show that explanations are a valuable tool both for improving our understanding of ConvNets' predictions and for designing more reliable models.