DOCTOR: A Simple Method for Detecting Misclassification Errors
arXiv:2106.02395
Abstract
Deep neural networks (DNNs) have shown to perform very well on large scale object recognition problems and lead to widespread use for real-world applications, including situations where DNN are implemented as "black boxes". A promising approach to secure their use is to accept decisions that are likely to be correct while discarding the others. In this work, we propose DOCTOR, a simple method that aims to identify whether the prediction of a DNN classifier should (or should not) be trusted so that, consequently, it would be possible to accept it or to reject it. Two scenarios are investigated: Totally Black Box (TBB) where only the soft-predictions are available and Partially Black Box (PBB) where gradient-propagation to perform input pre-processing is allowed. Empirically, we show that DOCTOR outperforms all state-of-the-art methods on various well-known images and sentiment analysis datasets. In particular, we observe a reduction of up to of the false rejection rate (FRR) in the PBB scenario. DOCTOR can be applied to any pre-trained model, it does not require prior information about the underlying dataset and is as simple as the simplest available methods in the literature.
arXiv commentThis paper has been accepted to appear as a spotlight in the Proceedings of the 2021 Conference on Neural Information Processing Systems (NeurIPS 2021), December 6-14, 2021, Virtual Event
References in corpus (9)
- Distilling the Knowledge in a Neural Network
- Densely Connected Convolutional Networks
- Rényi Divergence and Kullback-Leibler Divergence
- A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
- A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks
- Learning Confidence for Out-of-Distribution Detection in Neural Networks
- SelectiveNet: A Deep Neural Network with an Integrated Reject Option
- Bias-Reduced Uncertainty Estimation for Deep Neural Classifiers
- Robust Out-of-distribution Detection for Neural Networks