Just rotate it! Uncertainty estimation in closed-source models via multiple queries
This addresses the challenge of uncertainty estimation for users of closed-source models, offering a practical solution with demonstrated gains, though it is incremental as it builds on existing transformation-based methods.
The paper tackles the problem of estimating uncertainty in closed-source deep neural network image classification models by creating multiple transformed versions of an input image and querying the model's top-1 predictions, resulting in significant improvements in calibration compared to a naive baseline.
We propose a simple and effective method to estimate the uncertainty of closed-source deep neural network image classification models. Given a base image, our method creates multiple transformed versions and uses them to query the top-1 prediction of the closed-source model. We demonstrate significant improvements in the calibration of uncertainty estimates compared to the naive baseline of assigning 100\% confidence to all predictions. While we initially explore Gaussian perturbations, our empirical findings indicate that natural transformations, such as rotations and elastic deformations, yield even better-calibrated predictions. Furthermore, through empirical results and a straightforward theoretical analysis, we elucidate the reasons behind the superior performance of natural transformations over Gaussian noise. Leveraging these insights, we propose a transfer learning approach that further improves our calibration results.