ConvNets and ImageNet Beyond Accuracy: Understanding Mistakes and Uncovering Biases
Preprint
Abstract
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.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Adversarial Examples in the Physical World2018
- Methods for interpreting and understanding deep neural networksDigital Signal Processing, 2018
- Analysis of classifiers’ robustness to adversarial perturbationsMachine Learning, 2017
- DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks2016
- Exploring the space of adversarial images2016
- DeeperCut: A Deeper, Stronger, and Faster Multi-person Pose Estimation Model2016
- Human Pose Estimation via Convolutional Part Heatmap Regression2016
- Corpus-based estimates of word association predict biases in judgment of word co-occurrence likelihoodCognitive Psychology, 2014
- Case-based reasoning in the health sciences: What's next?Artificial Intelligence in Medicine, 2006
- Human problem solving: The state of the theory in 1970.American Psychologist, 1971