Improving robustness of classifiers by training against live traffic
This addresses robustness issues for deployed classifiers in real-world applications, but it is incremental as it builds on prior regularization and GAN-based methods.
The paper tackles the problem of deep learning models being overconfident on out-of-distribution (OOD) inputs by proposing an adaptive regularization technique that leverages live traffic data, which contains both in-distribution and OOD samples, to improve OOD detection without degrading in-distribution performance. Experiments on natural image datasets show significant improvements in OOD detection.
Deep learning models are known to be overconfident in their predictions on out of distribution inputs. This is a challenge when a model is trained on a particular input dataset, but receives out of sample data when deployed in practice. Recently, there has been work on building classifiers that are robust to out of distribution samples by adding a regularization term that maximizes the entropy of the classifier output on out of distribution data. However, given the challenge that it is not always possible to obtain out of distribution samples, the authors suggest a GAN based alternative that is independent of specific knowledge of out of distribution samples. From this existing work, we also know that having access to the true out of sample distribution for regularization works significantly better than using samples from the GAN. In this paper, we make the following observation: in practice, the out of distribution samples are contained in the traffic that hits a deployed classifier. However, the traffic will also contain a unknown proportion of in-distribution samples. If the entropy over of all of the traffic data were to be naively maximized, this will hurt the classifier performance on in-distribution data. To effectively leverage this traffic data, we propose an adaptive regularization technique (based on the maximum predictive probability score of a sample) which penalizes out of distribution samples more heavily than in distribution samples in the incoming traffic. This ensures that the overall performance of the classifier does not degrade on in-distribution data, while detection of out-of-distribution samples is significantly improved by leveraging the unlabeled traffic data. We show the effectiveness of our method via experiments on natural image datasets.