Approaching Test Time Augmentation in the Context of Uncertainty Calibration for Deep Neural Networks
This work addresses the need for reliable uncertainty estimates in deep learning models for real-world applications, though it is incremental as it builds on existing test time augmentation and calibration methods.
The authors tackled the problem of improving uncertainty calibration in deep neural networks for image classification by proposing two novel test time augmentation techniques, M-ATTA and V-ATTA, which outperformed state-of-the-art post-hoc calibration methods on datasets like CIFAR-10 and CIFAR-100 without affecting accuracy.
With the rise of Deep Neural Networks, machine learning systems are nowadays ubiquitous in a number of real-world applications, which bears the need for highly reliable models. This requires a thorough look not only at the accuracy of such systems, but also at their predictive uncertainty. Hence, we propose a novel technique (with two different variations, named M-ATTA and V-ATTA) based on test time augmentation, to improve the uncertainty calibration of deep models for image classification. By leveraging na adaptive weighting system, M/V-ATTA improves uncertainty calibration without affecting the model's accuracy. The performance of these techniques is evaluated by considering diverse metrics related to uncertainty calibration, demonstrating their robustness. Empirical results, obtained on CIFAR-10, CIFAR-100, Aerial Image Dataset, as well as in two different scenarios under distribution-shift, indicate that the proposed methods outperform several state-of-the-art post-hoc calibration techniques. Furthermore, the methods proposed also show improvements in terms of predictive entropy on out-of-distribution samples. Code for M/V-ATTA available at: https://github.com/pedrormconde/MV-ATTA